KR100839881B1 - Packet transmission method and packet transmission device - Google Patents

Abstract

A packet transmission method of a network device for transmitting a packet is disclosed. In the packet transmission method, the apparatus on the transmitting side makes two copies from the transmission packet, assigns each of the copied packets a sequence number identifying the same transmission order, and further, each of the copied packets. An identifier according to the transmission / reception pair is transmitted, and the transmission / reception apparatus receives the packet by two receiving means, respectively, confirms the identifier according to the transmission / reception pair, and if the identifier matches, from the sequence number The packet including the same information and its order are identified, and one of the packets of the same order is selected and sent out downstream, and when the other packets are closed or arrive only on one side, the arrived packet is transmitted downstream.

A packet transmitting apparatus, a transmitting function means, a receiving function means, a copying means, a number / identifier assigning means, a packet transmitting means, a packet receiving means, a memory, and a selecting means.

Description

Packet transmission method and packet transmission device {PACKET TRANSMISSION METHOD AND PACKET TRANSMISSION DEVICE}

The present invention relates to a packet transmission method and a packet transmission device, and more particularly, to a packet having a variable length such as an Ethernet transmission method, an MPLS transmission method, an GRE encapsulating, an IPinIP encapsulation, a GFP encapsulation, and a layer 2 frame encapsulation. The present invention relates to a packet transmission method and a packet transmission apparatus for realizing fault recovery without packet loss in a network employing an encapsulated transmission method, and the like.

Conventional fault detection and switching techniques in packet communications include, for example, Spanning Tree Protocol (STP) over Ethernet and derivatives thereof (see Non-Patent Documents 1 to 3, for example), or EAPS (Ethernet Automatic Protection). Switching). In addition, EAPS has a technology capable of switching at high speed in a ring topology for MAN service.

This technique selects a master switch in a ring, turns one side of the master switch into the primary switch, the other side into the secondary switch, and shuts off the secondary switch. The “Hello Packet” is sent from the main switch to the ring, and within a certain period of time, it does not return to the sub switch and detects the fault. In addition, the intermediate switch on the ring can transmit "TRAP" to the master switch when detecting the failure, and can detect the failure faster (less than 1 second). When a fault is detected, the sub switch is immediately set to "FORWARDING".

Further, in a ring-type network, RPR (Resilient Packet Ring) is a standard for realizing high reliability (see Non-Patent Document 4, for example).

In addition, the same information is transmitted to two or more lines on the ATM side by describing a cell number in the ATM cell as a non-disruptive technology of the ATM cell in the ATM line, copying two or more lines, and transmitting and transmitting them on different lines. There is a method of transmitting downstream by adopting one of the cells (see Patent Document 1, for example).

Non-Patent Document 1: IEEE 802.1D Spanning Tree Protocol (STP)

Non-Patent Document 2: IEEE 802.1w Rapid Spanning Tree (RSTP)

Non-Patent Document 3: Thorough Commentary on LAN Switching, Nikkei BT, ISBN4-8222-8099-3

Non-Patent Document 4: IEEE802.17 Resilient Packet Ring (RPR)

Patent Document 1: Japanese Patent Application Laid-Open No. 7-46250

Disclosure of the Invention

Challenges Solved by the Invention

However, the STP and EAPS of the above-described technique have a problem that high reliability that does not allow packet (frame) loss cannot be realized because the switching time is 1 second or more. In addition, this technique is a failure recovery technique in units of switch ports, and does not realize switching in units of packets. In addition, there is a problem that EAPS or RPR can not be applied other than the ring network.

Further, the non-interruptible technology for ATM cells described in Patent Document 1 is specialized for ATM lines, and because it is an uninterruptible technology for each line, there is a problem that it cannot be applied to node failure.

SUMMARY OF THE INVENTION The present invention has been made in view of the above, and provides a packet transmission method and a packet transmission apparatus for realizing high reliability failure recovery without packet loss in a network performing communication by transmitting a packet. The purpose.

Means to solve the problem

The above problem is a packet transmission method in a network apparatus for transmitting a packet.

The apparatus on the transmitting side creates two copies of the transmission packet, assigns each of the copied packets a sequence number for identifying the same transmission order, and further, an identifier according to the transmission / reception pair to each of the copied packets. To send,

The receiving device receives the packet at each of two receiving means,

Confirm an identifier according to the transmission / reception pair,

If the identifiers match, identify the packet and the sequence including the same information based on the sequence number,

One of the packets of the same order is selected to be sent downstream, and the other packets are discarded. When only one of the packets arrives, the received packet is transmitted downstream.

Moreover, this invention is a packet transmission method in the network apparatus which transmits a packet,

The apparatus on the transmitting side creates two copies of the transmission packet, assigns each of the copied packets a sequence number for identifying the same transmission order, and further, an identifier according to the transmission / reception pair to each of the copied packets. And assigns and transmits an identifier according to a path,

The apparatus at the transmitting side receives the packets at two receiving means, respectively,

Identifying an identifier according to the transmission / reception pair and an identifier according to the path,

If the identifiers match, the packets including the same information and their order are identified based on the sequence numbers.

It is also possible to configure as a packet transmission method, wherein one of the packets of the same order is selected and sent out downstream, and the other packets are discarded.

In the apparatus on the receiving side, it is possible to temporarily store packets transmitted in two paths in two FIFO memories, and select packets transmitted normally and transmit them downstream. In addition, the configuration may be configured to temporarily store packets transmitted in two paths in two circular hash memories, and select packets transmitted normally in a downstream manner, and temporarily transmit packets transmitted in two paths in two FIFO memories. The packet transmitted normally may be selected and transmitted downstream by using the third memory shared by the two receiving means as a cyclic hash.

Moreover, this invention is a packet transmission apparatus which transmits a packet,

Copying means for making two copies of a transmission packet;

Number / identifier assigning means for assigning a sequence number for identifying the same transmission order to each of the packets copied by the copying means, and further assigning an identifier according to the transmission / reception pair to each of the copied packets; And

Transmission function means for transmitting two packets to which the sequence number and the identifier are assigned;

Packet receiving means for receiving each of the two packets transmitted from the transmitting function means;

A memory for storing each of the two received packets;

Read the two packets stored in the memory, recognize the identifier according to the transmission / reception pair, and if the identifiers match, identify the packet including the same information and the sequence based on the sequence number, Selection means for selecting one of; And

A packet transmitting means comprising: receiving means for transmitting the packet selected by the selecting means downstream, discarding other packets, and transmitting the received packet downstream if only one piece arrives. It can also be configured as a device.

Moreover, this invention is a packet transmission apparatus which transmits a packet,

Copying means for making two copies of a transmission packet;

A sequence number for identifying the same transmission order is assigned to each of the packets copied by the copying means, and further, a number and identifier for assigning an identifier according to a transmission / reception pair and an identifier along a path to each of the copied packets. Grant means;

Transmission function means for transmitting two packets to which the sequence number and the identifier are assigned;

Packet receiving means for receiving each of the two packets transmitted from the transmitting function means;

A memory for storing each of the two received packets;

Read the two packets stored in the memory, recognize the identifier according to the transmission / reception pair and the identifier according to the path, and if the identifiers match, the packet having the same information based on the sequence number Selecting means for identifying a sequence and a sequence thereof, and selecting one of the packets of the same sequence; And

A packet transmission comprising a receiving function means comprising: transmitting the packet selected by the selecting means downstream, discarding another packet, and transmitting the received packet downstream when only one piece arrives. You may comprise as an apparatus.

In addition, the present invention provides a packet transmission method executed by a packet transmission apparatus provided at a packet transmission side and a reception side in a communication network for transmitting a packet by determining a transmission destination with reference to packet destination information.

Two or more independent paths are established between the packet transmitter on the sending side and the packet transmitter on the receiving side.

The packet transmission apparatus on the transmitting side inserts information for distinguishing the order of packets in a portion of the packet which is not referred to in determining the destination of the packet, and copies the packet to generate two or more packets, Send each packet to the independent path,

The packet transmission apparatus on the receiving side receives the packets via the independent paths respectively, and refers to the information for distinguishing the order of each packet, thereby identifying the packet and the order containing the same information, and identifying the same information. It is also possible to configure a packet transmission method, characterized in that one of the packets included is transmitted downstream in the order of the packets.

In addition, the present invention provides a plurality of packet transmission apparatuses provided in two or more independent paths on a packet transmission side and a reception side in a communication network for transmitting a packet by determining a transmission destination with reference to packet destination information. The packet transmission apparatus used in the packet transmission system comprising a,

The packet transmission apparatus includes a transmission function means and a reception function means,

The transmission function means,

Inserting means for inserting information for distinguishing the order of packets into a portion of the packet which is not referred to in determining a transmission destination of the packet;

Copying means for copying the packet to generate two or more packets; And

Sending means for sending out the respective packets to the independent paths; Equipped with

The receiving function means,

Receiving means for receiving packets each on the independent path;

Identification means for identifying a packet containing the same information and its order by referring to the information for distinguishing the order of each packet;

And a selecting means for transmitting one of the packets including the same information identified by the identifying means downstream in the order of the packets, in a packet transmission apparatus.

Further, the present invention is a packet transmission method executed by a packet transmission apparatus provided on a packet transmission side and a reception side in a communication network for transmitting a packet by determining a transmission destination with reference to packet destination information.

Two or more independent paths are established between the packet transmitter on the sending side and the packet transmitter on the receiving side.

By referring to a part of the packet header of the packet, the packet transmission apparatus on the transmitting side distinguishes between a high-reliability packet and a packet that is not, transmits the high-reliability packet by copying it to all of the independent paths,

The packet transmission apparatus on the receiving side determines whether or not the packet has arrived in the independent path by referring to a part of the packet header, and determines whether the packet has arrived in a plurality of paths. If the same packet arrives in two or more paths, only one is sent downstream and the rest is discarded. If the same packet arrives in only one path, the packet is sent downstream. The packet transmission method may be configured. By referring to a part of the packet header, it is possible to determine the priority of the packet, thereby determining whether or not it is a high reliability packet.

If the packet to be transmitted is an Ethernet packet, a part of the packet header may include a port number arriving at a switch in front of the packet transmitter, a type value of a layer 3 protocol in the frame header, a destination MAC address in the frame header, Source MAC address, priority (CoS value) included in the 802.1Q VLAN tag, VLAN-ID, DiffServ code point value (ToS value) included in the IP header, destination port number of UDP, source port number, TCP destination Is either a port number or a source port number.

When the packet to be transmitted is an MPLS-compatible packet, part of the packet header is any one of a destination MAC address, a source MAC address, and a CoS value (Exp value) of a shim header.

When the packet to be transmitted includes an IP packet, part of the packet header is any one of a ToS value, a source IP address, and a destination IP address of the IP packet.

Further, the packet transmission apparatus on the receiving side can determine the identity of the packets arriving in the plurality of paths based on the values obtained by operating a predetermined function with respect to the packets arriving in the plurality of paths.

In addition, the packet transmission apparatus on the transmission side inserts a sequence identifier or a timestamp into a packet to be transmitted, and the packet transmission apparatus on the reception side refers to the sequence identifier or timestamp inserted in the transmission side, thereby providing a plurality of paths. The identity of the packet arriving at may be determined.

In addition, the format of the sequence identifier or timestamp inserted into the packet by the packet transmission apparatus on the transmission side is in the same format as the VLAN tag of the 802.1Q specification. Order information or time information may be described.

Further, the present invention is a packet transmission method executed by a packet transmission apparatus provided on a packet transmission side and a reception side in a communication network for transmitting a packet by determining a transmission destination with reference to packet destination information.

Two or more independent paths are established between the sending packet transmitter and the receiving packet transmitter.

The packet transmission apparatus on the transmission side duplicates the packet and sends it out to all of the independent paths,

The packet transmission apparatus on the receiving side receives the packets via the independent paths respectively, and identifies the packets containing the same information by referring to the identification identification information of each packet, and any of the packets containing the same information. In the packet transmission method for transmitting an unsent packet downstream,

The packet transmission apparatus on the receiving side stores the identity identification information of the packet which has already been transmitted downstream for the packet preceding m pieces (m is an integer greater than or equal to 1) in the latest packet, and stores the identification identification information stored next. The packet transmission method may be configured by comparing the identification identification information of the arrived packets to determine whether the received packets have finished or not transmitted.

The identity identification information is a value obtained by acting a sequence identifier or time stamp inserted into a packet or a predetermined function on the packet.

Further, the present invention is a packet transmission method executed by a packet transmission apparatus provided on a packet transmission side and a reception side in a communication network for transmitting a packet by determining a transmission destination with reference to packet destination information.

Two or more independent paths are established between the packet transmitter on the sending side and the packet transmitter on the receiving side.

The packet transmission apparatus on the transmitting side inserts an order identifier for identifying the order of packets in the packet to be transmitted, duplicates the packet, and sends it out to all of the independent paths,

The packet transmission apparatus on the receiving side compares the value CF of the sequence identifier of the transmitted packet with the values of the sequence identifiers of the plurality of packets received on the independent paths, and receives the plurality of packets received on the independent paths. At least one packet of the sequence identifier of the packet that is greater than the value CF of the sequence identifier of the already transmitted packet and has the smallest value among the sequence identifiers of the plurality of packets received on the independent path. It is also possible to configure a packet transmission method, characterized in that the transmission of one of the packets downstream.

Further, the present invention is a packet transmission method executed by a packet transmission apparatus provided on a packet transmission side and a reception side in a communication network for transmitting a packet by determining a transmission destination with reference to packet destination information.

A plurality of independent paths are set between the sending packet transmitter and the receiving packet transmitter;

The packet transmitter on the transmitting side inserts a sequence identifier for identifying the order of packets in the packet to be transmitted, duplicates the packet, and sends it out to all of the independent paths,

The packet transmission apparatus on the receiving side receives the packets on the independent paths respectively and references the sequence identifiers of the respective packets to identify packets containing the same information and the sequence thereof, and the packets containing the same information. As a packet transmission method for transmitting one of the downstream in the order of the packet,

In the packet transmission apparatus on the receiving side, among the independent paths, the path in which the packet arrived first after the start of communication is the current system, and the value (CF) of the sequence identifier of the transmitted packet is received in the current system. By comparing the value of the sequence identifier of the packet to be transmitted, the packet having the sequence identifier larger than the value (CF) of the sequence identifier of the transmitted packet is designated as the next packet to be transmitted.

In the case where the arrival of a packet of the current system is interrupted for a predetermined time, another packet may be employed as a new current system, and a packet received by the system may be transmitted downstream. .

In addition, the present invention provides a plurality of packet transmission apparatuses provided through two or more independent paths on a packet transmitting side and a receiving side in a communication network for transmitting a packet by determining a transmission destination with reference to packet destination information. A packet transmission apparatus for use in a packet transmission system comprising:

The packet transmission apparatus includes a transmission function means and a reception function means,

The transmission function means,

By referring to a portion of the packet header of the packet, the means for distinguishing between a highly-reliable packet and a non-trusted packet, duplicating the high-reliable packet, and sending it to all of the independent paths,

The receiving function means,

With respect to the packet arriving on the independent path, it is determined whether or not it is a high reliability packet by referring to a part of the packet header, and for the high reliability packet, the sameness of the packet data arriving on a plurality of paths is determined, and the same packet is Means for transmitting only one downstream and discarding the other when arriving on two or more paths, and transmitting the packet downstream when the same packet arrives on only one path. It can be configured as a transmission device.

In addition, the present invention provides a plurality of packet transmission apparatuses provided in two or more independent paths on a packet transmission side and a reception side in a communication network for transmitting a packet by determining a transmission destination with reference to packet destination information. The packet transmission apparatus used in the packet transmission system comprising a,

The packet transmission apparatus includes a transmission function means and a reception function means,

The transmission function means includes means for replicating a packet and sending it out in all of the independent paths,

The receiving function means,

Means for receiving packets each in the independent path;

Means for identifying a packet containing the same information by referring to the identity identification information of each packet;

The identity identification information of the packet already transmitted downstream is stored for the packet preceding m (m is an integer of 1 or more) in the latest packet, and the stored identity identification information is compared with the identity identification information of the next arrival packet. Means for determining whether the corresponding arrival packet is complete or untransmitted;

One of the packets containing the same information, characterized in that it comprises a means for transmitting an untransmitted packet downstream, it may be configured as a packet transmission apparatus.

In addition, the present invention provides a plurality of packet transmission apparatuses provided in two or more independent paths on a packet transmission side and a reception side in a communication network for transmitting a packet by determining a transmission destination with reference to packet destination information. The packet transmission apparatus used in the packet transmission system comprising a,

The packet transmission device includes a transmission function means and a reception function means,

The transmitting function means includes means for inserting a sequence identifier for identifying a sequence of packets in a packet to be transmitted, duplicating the packet, and sending it out in all of the independent paths,

The reception function means compares a value CF of a sequence identifier of a transmitted packet with values of sequence identifiers of a plurality of packets received on the independent path, and compares the values of the plurality of packets received on the independent path. Of the at least one packet having a value greater than the value of the sequence identifier of the already transmitted packet (CF) among the value of the sequence identifier and having the smallest value among the sequence identifiers of the plurality of packets received on the independent path. It can also be configured as a packet transmission apparatus, characterized in that it comprises means for transmitting one packet downstream.

Further, in a communication network for transmitting a packet by determining a transmission destination by referring to packet destination information, a packet including a plurality of packet transmission apparatuses provided through at least two independent paths on the packet transmitting side and the receiving side. The packet transmission apparatus used in the transmission system,

The packet transmission apparatus includes a transmission function means and a reception function means,

The transmitting function means includes means for inserting a sequence identifier for identifying the order of packets in a packet to be transmitted, replicating the packet and sending it out through all of the independent paths,

The receiving function means receives the packets in the independent paths respectively, and refers to the sequence identifiers of the respective packets, thereby identifying a packet containing the same information and the sequence thereof, and selecting one of the packets containing the same information. Means for sending downstream in order of the packet,

The reception function means uses the current path as the path in which the packet arrives first after starting communication among the independent paths, the value CF of the sequence identifier of the transmitted packet, and the packet received in the current system. Compares the value of the sequence identifier of, and sets the packet having the sequence identifier larger than the sequence identifier value CF of the transmitted packet as the next packet to be transmitted.

In the case where the arrival of a packet in the current system is interrupted for a certain time, another system may be employed as a new current system, and the packet received by the system may be transmitted downstream. have.

Effects of the Invention

According to the present invention, in a packet network which could not be recovered in the event of a failure in the past, a copy of a packet is made, each of which is transmitted through a different path, and at the same time, the receiving side transmits one of the two. Therefore, a high reliability packet network can be provided by realizing the non-stop switching function. In addition, by using an independent path between transmission and reception, it is possible to provide a packet network of the same high reliability as point-point even in communication between a plurality of bases.

In addition, according to the present invention, since the specification of the packet requiring high reliability and the packet not requiring high reliability is performed with reference to the information of the packet header, the need for reliability can be used for various users' networks.

In addition, when the receiver determines the identity of the packet, by using a value that has a predetermined function applied to the packet, it is possible to realize high reliability of the network without inserting an excessive field into the packet.

Also, in the case where a packet is transmitted from one source to a plurality of destinations by performing the transmission after waiting for the arrival of the same packet in two or more paths on the receiving side, even when the sequence identifier is discontinuous, The skip of the sequence due to the delay of arrival can be eliminated.

In addition, priority control can be performed in the relay network by assigning a path identifier according to user priority.

1 is a system schematic diagram according to an embodiment of the present invention.

2 is a system schematic diagram according to one embodiment of the present invention;

3 is a diagram illustrating a configuration of a packet according to an embodiment of the present invention.

4 is a diagram showing the configuration of a packet according to an embodiment of the present invention (Example 1 in the case of Ethernet)

5 is a diagram showing a configuration of a packet according to an embodiment of the present invention (Example 2 in the case of Ethernet)

6 is a diagram showing the configuration of a packet according to an embodiment of the present invention (an example in the case of MPLS).

7 is a diagram showing the configuration of a packet according to an embodiment of the present invention (another example of encapsulation)

8 is a block diagram of a packet transmission apparatus according to the first embodiment of the present invention.

9 is a block diagram of a packet transmission apparatus according to the first embodiment of the present invention.

10 is another example of a transmission function unit of the packet transmission apparatus according to the first embodiment of the present invention.

Fig. 11 is a flowchart (1) of processing procedures of the counter section and the selection section according to the first embodiment of the present invention.

Fig. 12 is a flowchart (2) of the processing procedures of the counter section and the selection section according to the first embodiment of the present invention.

13 is another example of a packet transmission apparatus according to the first embodiment of the present invention.

Fig. 14 is a view for explaining the case where a circular hash is used in the memory according to the second embodiment of the present invention.

Fig. 15 is a flowchart of the processing procedures of the counter section and the selection section according to the second embodiment of the present invention.

16 is a block diagram of a packet transmission apparatus according to the third embodiment of the present invention.

17 is a block diagram of a packet transmission apparatus according to the third embodiment of the present invention.

Fig. 18 is a flowchart showing a procedure of writing operations from memory-A and memory-B to shared memory-C according to the third embodiment of the present invention.

Fig. 19 is a flowchart showing a data transfer procedure of shared memory C according to the third embodiment of the present invention.

20 is a diagram illustrating a configuration of overlapping an Ethernet packet according to a fourth embodiment of the present invention.

Fig. 21 is a diagram showing a configuration of overlapping Ethernet packets according to a fifth embodiment of the present invention.

Fig. 22 is a diagram for explaining multi-pointization by timeout setting according to the sixth embodiment of the present invention.

Fig. 23 is a diagram showing a configuration for assigning a VLAN tag according to the seventh embodiment of the present invention.

Fig. 24 is a view for explaining provision of a VLAN tag depending on a transmission path and a VLAN tag depending on a transmission source according to the seventh embodiment of the present invention.

Fig. 25 is a view for explaining provision of a VLAN tag depending on a transmission path and an ID depending on a transmission source according to the eighth embodiment of the present invention.

Fig. 26 is a diagram showing an application example for a network according to the ninth embodiment of the present invention.

27 is a diagram showing an application example to a network according to a tenth embodiment of the present invention.

28 is an example of planned uninterruptible conversion using LSS according to the eleventh embodiment of the present invention.

29 is an example of a planned non-stop transition using an LSS according to a twelfth embodiment of the present invention.

30 is a configuration diagram of a transmission function unit of a packet transmission apparatus according to a fourteenth embodiment of the present invention.

Fig. 31 is a block diagram of a transmission function unit of a packet transmission device according to a fourteenth embodiment of the present invention.

32 is a diagram showing an example where the priority of the user packet is reflected in the path identifier to be inserted.

33 is a diagram showing an example where priority is reflected in a counter field.

34 illustrates an example of inserting a counter inside a VLAN tag.

35 is a system configuration diagram according to a fifteenth embodiment of the present invention.

Fig. 36 is a diagram showing an example of the configuration of a transmission function unit of a packet transmitting apparatus according to a fifteenth embodiment of the present invention.

Fig. 37 is an example packet configuration when the target network is Ethernet.

38 is a block diagram of a reception function unit of a packet transmission apparatus according to a fifteenth embodiment of the present invention.

Fig. 39 is a flowchart showing the processing procedure of the reception function unit of the packet transmission apparatus according to the fifteenth embodiment of the present invention.

Fig. 40 is a flowchart showing the processing procedure of the reception function unit of the packet transmission apparatus according to the fifteenth embodiment of the present invention.

Fig. 41 is a block diagram of a receiving function unit of a packet transmitting apparatus according to the sixteenth embodiment of the present invention.

Fig. 42 is a flowchart showing the processing procedure of the reception function unit of the packet transmission apparatus according to the sixteenth embodiment of the present invention.

FIG. 43 is a diagram to help explain a memory that stores a timestamp; FIG.

44 is a diagram for explaining a memory that stores a function value.

45 is a flowchart showing a processing procedure of a reception function unit of a packet transmission apparatus according to the seventeenth embodiment of the present invention.

Fig. 46 is a flowchart showing the processing procedure of the reception function unit of the packet transmission apparatus according to the seventeenth embodiment of the present invention.

Fig. 47 is a block diagram of a reception function unit of a packet transmission apparatus according to the eighteenth embodiment of the present invention.

Fig. 48 is a flowchart showing the processing procedure of the reception function unit of the packet transmission apparatus according to the eighteenth embodiment of the present invention.

Fig. 49 is a flowchart showing the processing procedure of the reception function unit of the packet transmission apparatus according to the eighteenth embodiment of the present invention.

50 is a network configuration diagram according to a nineteenth embodiment of the present invention.

Fig. 51 is an example of a packet configuration of Ethernet according to the nineteenth embodiment of the present invention.

Fig. 52 is a block diagram of a reception function unit of a packet transmission apparatus according to the nineteenth embodiment of the present invention.

Fig. 53 is a block diagram of a receiving function unit of a packet transmitting apparatus according to a twentieth embodiment of the present invention.

Fig. 54 is a block diagram of a reception function unit of the packet transmission apparatus according to the twenty-first embodiment of the present invention.

Fig. 55 is a view for explaining the operation overview of the reception function unit of the packet transmission apparatus according to the twenty-first embodiment of the present invention.

Fig.56 is an example packet configuration of Ethernet in the twenty-first embodiment of the present invention.

[Description of the code]

100, 200, 300, 400, 500, 600, 70, 800 packet transmitter

110, 210 with transmission function

111, 211 receiver

115, 215 route / transmission pair discrimination unit

112, 212, copy

113, 213 Identifier

114, 214 transmitter

116 High Reliability Determination

120, 220, 320, 420, 520, 620, 720, 820 with receiving function

121, 221, 321, 421, 521, 621, 721, 821 receiver

122, 222, 322, 422, 522, 622, 722, 822 Memory A, Memory B, Memory C

123, 223 optional

124, 224, 427, 627, 727, 827 counter

125, 225, 325, 425, 525, 625, 725, 825 transmitter

126, 226 Identifier Reference

127, 227, 326, 426, 526, 626, 628, 826

829 arrival counter / weekly counter part

Implement the invention  Best form for

EMBODIMENT OF THE INVENTION Hereinafter, embodiment of this invention is described with drawing. 1 shows a schematic diagram of a packet transmission system according to an embodiment of the present invention.

In FIG. 1, the user networks 1 and 2 are connected via the wide area network 3. At the boundary between the user networks 1 and 2 and the wide area network 3, packet transmission apparatuses 4 and 5 according to the present invention are provided.

The packet transmission apparatuses 4 and 5 make two copies of the packets received from the user network 1 or 2 when transmitting the packets to the wide area network 3, and sequence numbers (counters) indicating the order in the packets. Value, also known as an order identifier), to perform transmission separately through independent paths. The receiving side selects one of a plurality of packets having the same information by referring to the counter value and transmits it downstream.

In the present specification and claims, "two or more independent paths" refers to other paths in which communication is not simultaneously interrupted by a single failure or construction. In addition, in the present specification and claims, the term "packet" is used to encompass the meaning of "frame" transmitted in a Layer 2 network.

In the configuration of Fig. 1, there may be a plurality of packet transmitting apparatuses on the transmitting side and a plurality of packet transmitting apparatuses on the receiving side. In that case, in the packet transmission apparatus 4, the packet transmission apparatus 5 is added to the packet by adding the identifier for identifying the transmission / reception pair of the packet (pair of the packet transmission apparatus on the transmission side and the packet transmission apparatus on the reception side) to the packet. In the above, packet selection may be performed for each transmission / reception pair. In the packet transmission device 4, it is also possible to assign a path identifier indicating a transmission path in the wide area network 3 and to transmit the packet along the path according to the identifier. It is also possible to add identification information that combines identification information of the transmission / reception pair (or only the destination or only the transmission source) and identification information of the path.

The user network and the wide area network are, for example, Ethernet, MPLS, a network that performs packet transmission by performing other encapsulation, and the like. Further, the wide area network is a network composed of a layer 2 switch or the like for transmitting a packet by determining a transmission source with reference to the packet destination information. In addition, it is preferable that a counter value used for identifying a packet having the same information is inserted in a portion which is not referred to for destination determination by the layer 2 switch or the like during packet transmission.

2 shows an example of a network configuration when the user network and the wide area network are Ethernet networks.

In FIG. 2, the Ethernet networks 1 and 2 at both ends are connected using the intermediate Ethernet network 3. Ethernet switches 6 and 7 are provided at the boundary of the Ethernet network 3. The packet transmission apparatuses 4 and 5 used in the present invention are provided outside the Ethernet switches 6 and 7 and are used to construct a highly reliable Ethernet network. The packet transmission apparatuses 4 and 5 create two copies of the Ethernet packets received by the Ethernet network 1 or 2 and transmit them. The transmission is performed individually, and the receiving side selects a normal one and transmits it downstream. As shown in Fig. 3, the copied packet is a packet in which two fields, which are a tag area and a counter area, are added in addition to the normal Ethernet header and payload, between the packet transmission apparatuses 4 and 5; Is sent from.

For example, adding a field may be performed as follows.

If the target packet is an Ethernet packet, a tag field and a counter field are inserted after the source MAC address of the Ethernet packet, and a VLAN tag and a sequence number along the path are described. It is also possible to insert a tag field and a counter field following the sender MAC address and describe the VLAN tag and sequence number according to the transmission / reception pairs. It is also possible to insert a tag field and a counter field following the sender MAC address, and describe the VLAN tag and sequence number according to the transmission / reception pair and the transmission path. The tag field and the counter field may be inserted after the sender MAC address, and the VLAN tag according to the transmission path, the identification ID and the sequence number according to the transmission / reception pair may be described.

In the case of using MPLS as a packet transmission technique, a tag field and a counter field may be inserted before a shim header of the MPLS, and a shim header and a sequence number according to a transmission path may be described. The tag field and the counter field may be inserted before the shim header of the MPLS to describe the shim header and the sequence number according to the transmission / reception pairs. In addition, the tag field and the counter field may be inserted before the shim header of the MPLS, and the shim header according to the transmission path, the identification ID and the sequence number according to the transmission / reception pair may be described, and the tag field and the counter field before the shim header of the MPLS. And insert a shim header and a sequence number according to a transmission path and a transmission / reception pair.

In addition, as a packet transmission technique, when encapsulation of variable length packets is used, a counter field is inserted in the rear side of the header for encapsulation, and a sequence number is described. In the receiving device, the identifier corresponding to the transmission / reception pair or the identifier corresponding to the path can be extracted from the header for encapsulation.

The sequence of packets corresponding to each system such as Ethernet and MPLS will be described in detail below.

4 is a diagram showing an example of a packet configuration in the case where the packet transmission apparatus of the present invention is applied to Ethernet.

(a) and (b) are normal packets transmitted from the user side. As shown in (a), the layer 2 includes a layer 2 (L2) header area (including a MAC address and the like) and a data (payload) area. (b) shows the case where the VLAN tag (user VLAN) used by the user side is included. Also in each case described below, an example including and not including a user VLAN is shown. However, since only the presence or absence of a user VLAN differs from an example including and without a user VLAN, a user VALN is selected. Only examples that are not included will be described. In addition, a plurality of user VALN tags may be provided.

The packet shown in (c) includes an area of a LAN segment identifier (for example, a VLAN identifier) showing a LAN segment of a wide area network to which a packet is transmitted, an area of an identifier of a transmission / reception pair, in an area of a VLAN tag, It includes a counter area. The transmission / reception pair identifier can be determined by, for example, MAC addresses of the source packet transmitter and the destination packet transmitter.

Here, when there is one transmission / reception pair or when the transmission / reception pair is not distinguished, the identifier of the transmission / reception pair may be omitted (e).

In the case where a plurality of copied packets are each transmitted to another network, the LAN segment identifier can be omitted as shown in (g) because it is not necessary to identify a path in the network. Furthermore, when there is only one transmission / reception pair or when the transmission / reception pairs are not distinguished, the configuration may include only the counter area as shown in (i).

5 is a diagram showing another example of the packet configuration in the case where the packet transmission apparatus is applied to Ethernet. An example of (a) includes an area of an identifier for identifying a LAN segment to which a packet is transmitted and a destination packet transmission device, an area of an identifier for identifying a source packet transmission device, and an area of a counter. In addition, (c) is an example including the area | region of an identifier for identifying a LAN segment and a transmission / reception pair, and the area | region of a counter.

Fig. 6 is a diagram showing an example of a packet structure in the case where the packet transmission device is applied to MPLS. (a) shows a normal packet transmitted from the user side. In addition, (b) shows a normal packet in Ethernet over MPLS etc. (b) differs from (a) in that the L2 header is included behind the L3 header. In each case of FIGS. 6 and 7, an example in which the L2 header is not included in the rear side of the L3 header and an example in which the L2 header is included are shown. Since only the presence or absence of such an L2 header differs, only an example in which the L2 header is not included behind the L3 header will be described.

6 (c) is provided with a transmission / reception pair area and a counter area in front of the shim header. (E) omits the area of the transmission / reception pair in (c).

FIG. 7 is a diagram showing an example of a packet configuration when the packet transmission device is applied to a technique of encapsulating and transmitting a packet. (a) and (b) are normal packets coming out from the user side.

(c) includes an identifier area for identifying a path for transmitting a wide area network, an identifier area for a transmission / reception header, and a counter area before the encapsulation header. In addition, when it is not necessary to identify a transmission / reception pair, it can be set as (e).

In addition, an identifier indicating a path such as a LAN segment and an identifier indicating a transmission destination, a transmission source, a transmission / reception pair, or the like may be referred to as a path identifier.

As shown in Figs. 4-7, the area of the counter is preferably provided in a portion which is not referenced for destination determination during the transmission of the packet. Thereby, there is an advantage that the operation of the packet transmission apparatus of the present invention can be realized without depending on the type of carrier of the wide area network in which the packet is transmitted. In this method, it is preferable to insert into a packet and to refer to an identifier referred to during transmission for destination determination in the case of Ethernet, in the form of a VLAN tag in consideration of the affinity for the network. On the other hand, the format of the counter area is similarly preferably in the form of a VLAN tag, but may be any field length in consideration of not referencing during transmission. Also, in the case of other than Ethernet, the length of the counter area may be arbitrary. Hereinafter, each embodiment is described.

[First embodiment]

8 is a diagram showing the configuration of a packet transmission apparatus 100 that realizes the first embodiment of the present invention.

The packet transmission apparatus 100 illustrated in FIG. 8 includes a transmission function unit 110 and a reception function unit 120. When the packet transmission apparatus 100 is the transmission side, the transmission function unit 110 is used, and when the packet transmission apparatus 100 is the reception side, the reception function unit 120 is used.

The transmission function unit 110 includes a receiver 111, a copy unit 112, an identifier granter 113 ₁ , 113 ₂ , and a transmitter 114 ₁ , 114 ₂ .

The transmission function unit 110 receives the packet from the user network side in the reception unit 111 and generates two copies of the packet in the copy unit 112. The copied packets are sent to the identifier assigning units 113 ₁ and 113 ₂ , where a counter area (order identifier area) is given to the packet, and a counter value (order identifier) is recorded in the area. The counter value indicates the order of the packets, and increments and records them by one when the packets are transmitted. The packets generated by the identifier assigning units 113 ₁ and 113 ₂ are transmitted to the wide area network side by the transmitting units 114 ₁ and 114 ₂ .

The reception function unit 120 includes a reception unit 121, a memory A 122 _A , a memory B 122 _B , a selection unit 123 ₁ , 123 ₂ , a counter unit 124, and a transmission unit 125. do.

In the reception function unit 120, two reception units 121 ₁ and 121 ₂ receive packets of the A-type and B-type, respectively, from the wide area network. The received packet is stored in the memory A 112 _{A in the A} system, and stored in the memory B 112 _{B in the} B system. In this embodiment, the memory A 112 _A and the memory B 112 _B are used as FIFOs (First In First Out), respectively. Packets read out from the memory A 112 _A and the memory B 112 _B are identified and selected by the selectors 123 ₁ and 123 ₂ , remove the counter area added to the packet, and send the packet to the user network. Transmitted by 125.

Fig. 9 is a diagram showing another configuration of the packet transmission device for realizing the first embodiment of the present invention. The example shown in FIG. 9 is an example in which there exists a plurality of transmission side packet transmission apparatuses and a destination side packet transmission apparatus, and identifies a transmission / reception pair.

The packet transmission apparatus 100 shown in FIG. 9 includes a transmission function unit 110 and a reception function unit 120. In Fig. 9, when the packet transmission device 100 is the transmission side, the transmission function unit 110 is used, and when the packet transmission device 5 is the reception side, the reception function unit 120 is used.

The transmission function unit 110 includes a receiver 111, a path / transmission pair discrimination unit 115, a copy unit 112, an identifier assigning unit 113 ₁ , 113 ₂ , and a transmitter 114 ₁ , 114 ₂ . have.

The transmission function unit 110 receives the packet from the user network side in the reception unit 111. Then, the path / transmission / pair pair determining unit 115 determines the packet transmission path and the transmission / reception pair (of the packet transmission device) from the packet destination or the like. In addition, when the path for transmitting the plurality of copied packets is determined, it is not necessary to determine the packet transmission path.

The copy 112 then generates two copies of the packet. The copied packet is sent to the identifier assigning units 113 ₁ and 113 ₂ , where an area of a path identifier (including path identification information and transmission / reception pair identification information) and a counter area are provided to the packet. Uh, the value is recorded in each area.

Examples of the path identifier area and the counter area are as shown in FIGS. 4 to 7. The sequence number (order identifier) is also recorded in the counter area. The packets generated by the identifier assigning units 113 ₁ and 113 ₂ are transmitted to the wide area network by the transmitting units 114 ₁ and 114 ₂ . In addition, the transmission function part 100 may be the structure shown in FIG. In this case, the packet to which the identifier and the counter value were given by the identifier provision part 113 is copied in the copy part 112. As shown in FIG.

The reception function unit 120 of FIG. 9 includes the reception units 121 ₁ and 121 ₂ , the memory A 122 _A , the memory B 122 _B , the identifier reference units 126 ₁ and 126 ₂ , the control unit 127, The selector 123 ₁ , 123 ₂ , the counter 124, and the transmitter 125 are configured. In addition, the selection unit and the counter unit are provided for each transmission source, and the selection units 123 ₁ and 123 ₂ and the counter unit 124 correspond to one of the transmission sources. In addition, when employing a plurality of different path pairs among the same transmission / reception pairs, the selection unit 123 ₁ , 123 ₂ and the counter unit 124 may be provided for each path pair of the same transmission source. This is the same in other embodiments.

In the reception function unit 120, the two reception units 121 ₁ and 121 ₂ receive packets of the A-type and the B-type from the wide area network 3, respectively. The received packet is stored in the memory A 122 _{A in the A} system and in the memory B 122 _{B in the} B system. In this embodiment, the memory A 122 _A and the memory B 122 _B are used as FIFOs (First In First Out), respectively. In the data read out from the memory A 122 _A and the memory B 122 _B , the path identifier is referred to in the identifier references 126 ₁ and 126 ₂ , and the source packet transmitter is identified.

And based on the control of the control part 127, a packet is transmitted to the selection part 123 ₁ , 123 ₂ corresponding to the transmission source of a packet, and is selected by the selection part 123 ₁ , 123 ₂ and the counter part 124. FIG. The packet is selected based on the sequence number, and the packet is transmitted by the packet transmitter 125 to the user network side by removing the packet's path identifier area and counter area.

Next, the processing procedure by the counter part 124 and the selection part 123 is demonstrated. This processing procedure is the same in FIGS. 8 and 9.

11 is a flowchart showing the processing procedures of the counter section and the selection section according to the first embodiment of the present invention. The following processing is processing corresponding to one transmission / reception pair. In the configuration shown in Fig. 9, the following processing is performed for each transmission / reception pair (per transmission source in the reception function section) by the functions of the identifier reference section and the control section.

In the transmission function unit 110 and the reception function unit 120, two systems exist. Each receiving unit waits for the arrival of a packet and first stores the received packet in a memory used as a FIFO. When the memory A with the packet selection unit (123 _1), with reference to the oldest packet is acquired (step 101, 102), the counter value of the packet in the FIFO, and that a CA. The CA and the reference counter CF managing the counter section are compared (step 103), and conditional branching is performed according to three cases, where CF = CA, CF> CA, and CF <CA. First, if CF = CA, the packet is transmitted to the packet transmitter 125, and the packet transitions to the packet waiting state (step 104). If CF> CA, the packet is discarded and transitioned to the packet waiting state (step 105). If CF <CA, the system transitions to the packet waiting state of system B (step 106).

If the packet to the B type, a selection unit (123 ₂₎ is obtained, the oldest packet in the memory B (122 _B) used as a FIFO of the B type (step 107) with reference to the counter value of the packet, and that a CB do. Here, CF and CB are compared (step 108), and conditional branching is performed according to three cases of CF> CB, CF = CB, and CF <CB. In the case of CF> CB, the packet is discarded and the packet transition state of the B system is reached (step 109). In the case of CF = CB, the packet is transmitted, and the value of CF is increased by one to transition to the packet waiting state of system A (step 110). In the case of CF <CB, the value of CF is added to 1 to make a transition to the packet waiting state of system A (step 111).

The packet sent to the packet transmitter 125 is removed from the path identifier area and the counter area and transmitted to the user network side.

By the above operation, it is realized that packets having the same information are identified in the order, and one of the packets having the same information is transmitted downstream in the order of the packets.

In the case of implementing the algorithm of the present embodiment, when the condition of CF = CA is continuous, there is a possibility that the memory B 122 _B of the B system is full. In this case, the following measures may be considered.

(1) The FIFOs of the memory A 122 _A and the memory B 122 _B are configured as cyclic FIFOs. In this configuration, the old packet is overwritten by the new packet without filling up the memory.

(2) When writing to the FIFOs of the memory A 122 _A and the memory B 122 _B , if a certain amount of data is exceeded, old data is forcibly removed.

(3) Regularly alternate the roles of the A system and the B system to prevent the memory B 122 _B of the B system from overflowing.

(4) Add to the algorithm. It shows the full algorithm steps are taken against that of the memory B (122 _B) for B based on Fig. The difference from FIG. 11 described above is that the condition of CF = CA is satisfied, the packet transmission is performed, and then the usage of the memory B 122 _B is checked, and if it exceeds the predetermined threshold value. In other words, the system transitions to the packet waiting state of system B (step 201). All other parts are the same as in FIG.

Further, the packet transmission device may be configured as shown in FIG. In the structure shown in FIG. 13 and the structure shown in FIG. 9, the structure of the receiving function part 120 differs. In the configuration of FIG. 9, the path identifier is distributed to the selection unit and the counter unit for each transmission / reception pair by referring to the path identifier as a packet read out from the memory. In the configuration of FIG. 13, the path identifier is first referred to in the reception packet for each transmission / reception pair. Store in memory. The operation of the configuration of FIG. 13 is performed in exactly the same manner as the operations shown in FIGS. 11 and 12 from the viewpoint of each transmission / reception pair.

Second Embodiment

In this embodiment, the apparatus structure uses what was shown in FIG. 8 or FIG. 13 in 1st Embodiment demonstrated above. In the first embodiment, the FIFO or the cyclic FIFO is used for the memory A 122 _A and the memory B 122 _B. However, in the present embodiment, the cyclic hash is used.

Fig. 14 is a diagram for explaining the case where the circular hash is used in the memory according to the second embodiment of the present invention.

In the circular hash, a given memory area is divided into n (integer), and 1 to n are given as addresses. It is assumed that a packet transmitted in a wide area network is stored in a memory area whose address is the remaining value when the counter value is divided by n with reference to the counter value. In the case of storing, if the payload area is included among the L2 header, the path identifier area, the counter area, and the data (payload) area, all may be stored, and some of them may be selected and stored. good. With respect to the contents of the memory A 122 _A and the memory B 122 _B configured in this way, the counter section and the selection section operate in the processing sequence shown in FIG.

15 is a flowchart of processing procedures of the counter section and the selection section according to the second embodiment of the present invention. In addition, the following process is a process corresponding to one transmission / reception pair. That is, by the functions of the identifier reference unit and the control unit, the following processing is performed for each transmission / reception pair (per transmission source in the transmission function unit).

In the transmission function unit 120, there are two systems, A system and B system, and a system for receiving packets is mainly referred to as a selective system and a preliminary system as a non-selective system. For example, in the case of mainly transmitting packets from the A system, the A system will be referred to as the selective system, and the B system will be referred to as the non-selective system. The distinction between the selective and non-selective systems is not fixed and may be changed depending on conditions. When packet processing is started, both systems of the A system and the B system enter a packet waiting state. Among them, the system that first receives the packet is a selection system. At this time, the read counter value is set to CF (step 301). Hereinafter, the A system is referred to as the selection system.

Next, the process proceeds to the main processing procedure.

According to the presence or absence of data of the place corresponding to the CF address of the memory A 122 _A of the selection system, the condition is branched into two (step 302).

(1) If there is data, the data of the CF address is taken out from the memory A 122 _A and transmitted in order to perform the transmission process (step 303). Thereafter, CF is increased by one (step 304), and again, the presence or absence of data at the place corresponding to the CF address of the memory A 122 _A of the selection system is checked.

(2) If no data has further two branch according to the data presence or absence of the memory B (122 _B) of the check whether the CF address data of the memory B (122 _B) of the non-selected system (Step 305, the non-selected-based .

(2-1) When there is data, and transmits the extracted data to present on the CF memory address of B (122 _B) of the non-selection system (step 306). Then, it is checked whether or not the data by increasing the 1 CF (step 304), selection-based memory A (122 _A) of the back.

(2-2) If there is no data in step 305, timeout wait is performed (step 307), and if the CF-th data arrives before the timeout (step 309), the above transmission processing of (2-1) is performed. In addition, a name change of the selection system and the non-selection system is performed (step 310). If the timeout occurs, the transmission is abandoned and the CF is increased by one (step 304), and the process proceeds to packet processing of the selection system.

By using the circular hash, the packet is stored in a memory area whose address is the remaining value when the counter value is divided by n. If a frame with a counter value of N reads the packet from the cyclic hash even when the counter arrives before the frame of Nn, the counter value is taken into account when the packet is read in the order of the counter values. Arrival order may be corrected in the correct order at the time of reading.

[Third Embodiment]

16 is a diagram showing the configuration of a packet transmission apparatus according to the third embodiment of the present invention.

The packet transmission device 200 shown in FIG. 16 includes a transmission function unit 210 and a reception function unit 200.

The transmission function unit 210 includes a receiver 211, a copy unit 212, an encoder 213 ₁ , 213 ₂ , and a transmitter 214 ₁ , 214 ₂ , and FIG. 8 of the first embodiment described above. It is the same configuration as shown in FIG.

The reception function unit 220 includes a reception unit 221 ₁ , 221 ₁ , a memory A 222 _A , a memory B 222 _B , a selection unit 223 ₁ , 223 ₂ , a counter unit 224, and a transmitter unit 225. ). The reception function unit 220 receives the packet from the reception units 221 ₁ and 221 ₂ , and records the packet in the memory A 222 _A or the memory B 222 _B. The memory A 222 _A and the memory B 222 _B are used as FIFOs.

The selectors 223 ₁ and 223 ₂ read data from the memory A 222 _A and the memory B 222 _B and transfer the data to the shared memory C 222 _C in the order described later.

The shared memory C 222 _C constitutes a circular hash shown in FIG. 14, is read out under the control of the counter unit 224, and is transmitted to the user network by the transmitter 225.

17 shows another example of the packet transmission apparatus 200 according to the present embodiment. The example shown in FIG. 17 is an example in which there exists a plurality of transmission side packet transmission apparatuses and a plurality of destination side packet transmission apparatuses, and identifies a transmission / reception pair.

In FIG. 17, the packet transmitter 200 includes a transmission function unit 210 and a reception function unit 220.

The transmission function unit 210 is a packet receiver 211, a path / transmission pair discrimination unit 215, a copy unit 212, an identifier grant unit 213 ₁ , 213 ₂ , and a transmitter 214 ₁ , 214 ₂ . It is comprised and is the same structure as shown in FIG. 9 in 1st Embodiment demonstrated above.

The reception function unit 220 includes a packet receiving unit 221 ₁ , 221 ₂ , a memory A 222 _A , a memory B 222 _B , a shared memory C 222 _C , and an identifier reference unit 226 ₁ , 226 ₂ . , A control unit 227, a selection unit 223 ₁ , 223 ₂ , a counter unit 224, and a transmission unit 225. The counter unit and the shared memory C are provided for each transmission source, and the counter unit 224 and the shared memory C 222 _C correspond to one of the transmission sources.

The reception function unit 220 receives the packet by the packet reception units 221 ₁ and 221 ₂ , and records the packet in the memory A 222 _A or the memory B 222 _B. Memory A 222 _A and memory B 222 _B serve as FIFOs.

The identifier references 226 ₁ and 226 ₂ refer to the path identifier of the packet and identify the source of transmission of the packet. And based on the control of the control part 227, the selection parts 223 ₁ and 223 ₂ store the data of the memory A 222 _A and the memory B 222 _B , and the identifier reference parts 226 ₁ and 226 ₂ . For each transmission source identified by, the transmission is performed to the shared memory C corresponding to the transmission source in the order described below.

The shared memory C 222 _C constitutes a cyclic hash shown in FIG. 14, and the packet is read out under the control of the counter unit 224, and is transmitted to the user network by the transmitting unit 225.

18 is a flowchart showing a recording procedure for the shared memory C in the memories A and B according to the third embodiment of the present invention. Hereinafter, although it demonstrates based on the structure of FIG. 17, the process is performed by the same operation also in FIG. 18 and 19 are processes corresponding to one transmission / reception pair. That is, the following processing is performed for each transmission / reception pair (per transmission source in the reception function unit) by the functions of the identifier reference unit and the control unit.

The identifier reference unit 226 and the selection unit 223 perform packet reception waiting (step 401), and if there is data in any one of the memory A 222 _A and the memory B 222 _B (step 402), This is obtained, and the sequence number is read out from the counter area of the packet (step 403). If the area of memory C 222 _C corresponding to the sequence number is not empty (step 404, NO), the packet is discarded (step 406), and if empty, the path identifier area and the counter area are removed to remove memory C (222 _C). (Step 406), and transition to packet reception wait.

19 is a flowchart showing the data transfer procedure of the shared memory C by the counter unit according to the third embodiment of the present invention.

In the packet waiting state (step 501), it is checked whether there is data in the F (integer) -th entry of the memory C (222 _C ) (step 502), and branches in the presence or absence of data.

(1) If there is data, it is transmitted to the transmitting unit 225 (step 503), and the value of F is increased by one (step 504) to return to the packet reception standby.

(2) If there is no data, wait for timeout. The condition branches to two further (step 505).

(2-1) If the F-th data arrives before the timeout, the data is transmitted (step 503), and the value of F is increased by one (step 504).

(2-2) In the case of timeout, it is checked whether an entry after F + 1 exists (step 506). Two additional branches based on the presence or absence of data.

(2-2-1) If an entry exists, the value of F is increased by one (step 504), and the packet transition state is reached.

(2-2-2) If the entry does not exist, the process transitions to the packet waiting state (step 501).

Fourth Embodiment

Next, the redundancy structure which concerns on 4th Embodiment is demonstrated. In the fourth to tenth embodiments, the case where the packet transfer apparatus of the present invention is applied over Ethernet will be described as an example. Incidentally, in the following description, the configuration of the transmitting device and the receiving device may be performed in any of the first to third embodiments described above, but in the present embodiment, it is necessary to identify the transmission / reception pairs. none.

Fig. 20A shows a configuration in which all Ethernet packets are duplicated. In the copying unit of the transmitting apparatus, two copies of the transmitting packet are created, and the identifier assigning unit newly assigns a VLAN tag ("VLAN-A" in Fig. 20A) and a sequence number for identifying the same transmission order. In addition, the transmitter transmits two copies to different networks. The selector of the transmitting-side apparatus selects one that arrived earlier among the packets of the same order and sends it downstream. Discard the other packet as it arrives.

FIG. 20 (B) shows a configuration in which the VLAN setting selects and duplicates only the Ethernet packet of "VALN-A". In the transmitting apparatus, the VLAN setting identifies a packet whose VLAN setting is VLAN-A, creates two copies of the packet in the copy unit, and newly assigns a VLAN tag (FIG. 20 (B)) to the identifier assigning unit. "VLAN-B") and a sequence number for identifying the same transmission sequence, and the packet transmitter transmits two copies to different networks, respectively.

[Fifth Embodiment]

21 is a diagram illustrating a configuration of overlapping the Ethernet packet according to the fifth embodiment of the present invention. In addition, in the following description, although the structure of a transmitter side apparatus and a receiver side apparatus may be performed by any of the structures of the 1st thru | or 3rd embodiment mentioned above, it is not necessary to identify a transmission / reception pair also in this embodiment. none.

21A illustrates a configuration in which a VLAN tag selects and duplicates an Ethernet packet of "VLAN-A". In the transmitting apparatus, two copies of the transmission packet are created in the copy unit, and the VLAN assignment unit ("VLAN-B" and "VLAN-C" in Fig. 21A) is used for each copy in the identifier assigning unit. Then, a sequence number for identifying the same transmission sequence is assigned, and the packet transmission unit transmits the same to the same network. In the receiving apparatus, two ports (receiving units) are configured to receive packets of "VLAN-B" and "VLAN-C", respectively.

In addition, when the "VLAN-B" packet and the "VLAN-C" packet are respectively transmitted in the wide area network, it is possible to use a separate and independent path without losing communication at the same time due to a single failure or construction work. have.

Fig. 21B is a configuration in which all packets to which a VLAN tag has not been assigned are duplicated. In the same way as in the upper stage, two copies of transmission packets copied from a copy unit in the transmitting apparatus are identifiers. In Fig. 21, a different VLAN tag (" VLAN-A " and " VLAN-B " in Fig. 21B) and a sequence number for identifying the same transmission order are newly assigned, and transmitted from the packet transmitter to the same network.

FIG. 21C is a configuration in which a VLAN tag selects and duplicates a packet of "VLAN-A" similarly to FIG. 21A. In the transmitting apparatus, two copies are respectively given different VLAN tags ("VLAN-B" and "VLAN-C" in Fig. 21C), and in the transmitting apparatus, one port setting is "VLAN-". The case where it is set to the structure which receives both B "and" VLAN-C "is shown. In this way, the settings of the transmitting apparatus and the receiving apparatus need not be identical.

[Sixth Embodiment]

Fig. 22 is a diagram for explaining multi-pointing according to the timeout setting according to the sixth embodiment of the present invention. In the following description, the configuration of the transmitting apparatus and the receiving apparatus is performed in the above-described first to third embodiments in a configuration capable of setting timeout.

In the example of Fig. 22, the reception section of the receiving side apparatus sets the time to wait for a packet of a certain sequence number, and when the predetermined time arrives in the transmission section of the receiving side apparatus, the next sequence number is automatically set. A timeout function is used to send the packet downstream. In the case of the present embodiment, a timer (not shown) for measuring a predetermined time is provided to the receiving unit and the transmitting unit of the receiving side apparatus.

[Seventh Embodiment]

Fig. 23 is a diagram illustrating a configuration for assigning a VLAN tag according to the seventh embodiment of the present invention.

Fig. 23A shows a configuration for assigning VLAN tags depending on the transmission / reception pairs. When copying two transmission packets in the copy unit of the transmitting apparatus and assigning the same sequence number in the identifier assigning unit, a VLAN tag depending on the transmitter pair is assigned, and the sequence number is set from 1 for each transmit / receive pair. The packet transmission unit transmits two copies to different networks. In this case, the packet corresponds to the remaining part of the path identifier of FIG. 5 (c) or (d) except for the part depending on the " LAN segment ". As shown in Fig. 23A, for example, a packet transmitted from the device X to the device P is given a VLAN tag called "VLAN-PX".

In the selection unit of the receiving device, the VLAN tag of the packet received from the two networks is checked, and for the packet including the VLAN tag indicating the same sender, the packet having the same information in the sequence number and the sequence thereof are identified. One of the packets of the same order is selected and sent by the transmitter, and others are discarded.

Regarding the VLAN tag, even if the VLAN packet is included or not included in the input packet, the identifier providing unit newly assigns the VLAN tag. This realizes multi-pointing.

In addition, the transmission / reception pair can be determined through, for example, MAC addresses of transmission sources and destinations of transmitted Ethernet packets, VLAN settings, and the like.

Fig. 23B is a configuration in which a VLAN tag is applied depending on a transceiver pair and a transmission path. In the copying section of the transmitting apparatus, when two transmission packets are copied and the identifier assigning unit assigns the same sequence number, the duplicated two packets have different VLAN tags, and each packet is different from each other. VLAN tag is assigned and sequence number is assigned from 1 for each transceiver pair. By doing so, it is possible to identify the path of two packets. Here, the packet corresponds to (c) or (d) of FIG. 5. In FIG. 23B, for example, VLAN tags called "VLAN-P1X" and "VLAN-P2X" are attached to packets sent from the device X to the device P. In FIG.

In the transmitting apparatus, by identifying the VLAN tag in the identifier reference section, the packet including the VLAN tag indicating the same transmission source is identified, and the packet having the same information and the sequence thereof are identified based on the sequence number. Accordingly, multi-pointing can be realized. For example, when a plurality of pairs of paths are used, packets that are the same transmission source and pass through the same path pair can be identified.

In addition, when assigning a VLAN tag according to a transmission / reception pair and a path, it is also possible to assign one VLAN tag corresponding to each. An example of such a case is shown in Figs. 24 (A) and (B). For example, in Fig. 24A, VLAN-P and VLAN-X are provided in place of VLAN-PX in Fig. 23A. In addition, in FIG. 24B, in place of VLAN-P1X in FIG. 23B, VLAN-P1 and VLAN-X are provided.

[Eighth Embodiment]

Fig. 25 is a view for explaining provision of an ID depending on a VLAN tag and a transmission / reception pair depending on the transmission path according to the eighth embodiment of the present invention. In the following description, the configuration of the transmitting apparatus and the receiving apparatus may be performed in any of the configurations of the first to third embodiments described above.

The example of FIG. 25 is a structure which gives a VLAN tag depending on a transmission path, and gives the identification ID which depended on a transceiver pair. When two copies of a transmission packet are copied from a copy unit of a transmitting device, and an identifier assigning unit assigns the same sequence number, a VLAN tag newly dependent on a transmission path is newly assigned so that the VLAN tags of two duplicated packets are different. Furthermore, different IDs and sequence numbers are assigned to each transceiver pair. Here, the packet generally corresponds to (c) or (d) of FIG. 7.

In the identifier reference section of the receiving device, a packet transmitted from the same source is identified through the same path pair with reference to the VLAN tag and ID, and a packet including the same information based on the sequence number and its sequence are identified. The transmitter sends one packet of the same sequence and discards the other packet.

[Ninth Embodiment]

Fig. 26 is a diagram showing an application example for a network according to the ninth embodiment of the present invention. In Fig. 26, an example applied to the metro loop is shown. In Fig. 26, " uninterruptible Ethernet " corresponds to the above-described packet transmission apparatus.

The configuration shown in Fig. 26 is a configuration in which the two packets copied by the above-described transmitting side apparatus are respectively transmitted in paths opposite to the ring configuration. For example, it can be used when transmitting a packet of the same VLAN tag to another network. In addition, another network in this embodiment is an example of a network which comprises a physically independent path | route.

[Tenth Embodiment]

27 is a diagram illustrating an application example to a network according to a tenth embodiment of the present invention. 27 shows an example applied to an Ethernet dedicated line. In Fig. 27, " non-stop Ethernet " corresponds to the above-described packet transfer apparatus.

The configuration shown in Fig. 27 is a configuration in which two packets copied by the transmitting apparatus are transmitted on different Ethernet dedicated lines. For example, it can be used when transmitting a packet of the same VLAN tag to another network. In addition, the other network which concerns on this embodiment is an example of the independent path | route using a dedicated line.

[Eleventh Embodiment]

28 shows an example of a planned non-stop conversion using LSS according to the eleventh embodiment of the present invention. In FIG. 28, an example of a case in which a packet transmission apparatus performs a non-stop switching intentionally using a Link Signaling Sublayer (LSS) protocol is shown. The counter is sent by preparing the LSS byte in the Inter Frame Gap (IFG) (Fig. 28 (A)), and measuring the counter delay difference between the current system and the standby system, thereby eliminating the delayed one. When the arrival of the counter of the current system is delayed due to a breakdown or the like, a packet of the preliminary system is adopted (Fig. 28 (B)).

[Twelfth Embodiment]

29A to 29C show examples of planned non-stop switching using the LSS according to the twelfth embodiment of the present invention. In Fig. 29, the packet transmission device switches between the current system and the standby system by sending a protocol such as APS (Automatic Protection Switching) inbound.

[Thirteenth Embodiment]

In this embodiment, the variable length encapsulation technique will be described.

As an encapsulation technique of a variable length packet, for example, it is possible to use an Internet encapsulation protocol. For example, a method of inserting a counter field between the capsule header and the datagram can be employed.

In addition, GRE encapsulation techniques can be used. The GRE encapsulation technique uses a method of attaching a transmission header + a GRE header to an original datagram as an encapsulation header. For example, a method of inserting a counter field between a GRE header and a datagram can be adopted. have.

In addition, IPinIP tunneling technology may be used. The IPinIP tunneling technique uses a method of attaching an external IP header + tunneling header to the original datagram as an encapsulation header, and a method of inserting a counter field between the tunneling header and the datagram can be adopted.

In addition, encapsulation by PPP or HDLC can be used. For example, a method of inserting a counter field between the PPP header and the datagram may be employed.

It is also possible to use encapsulation with GFP. For example, a method of inserting a counter field between the header and the datagram can be employed.

In addition, in the encapsulation technique, any technique capable of stacking between a header (overheader) and a datagram is applicable to the present invention.

Further, the packet transmission apparatus of the present invention can be realized by mounting a program that can execute the above-described transmission function and reception function units on a computer to be used as a packet transmission apparatus. The program can be distributed through a network.

The constructed program can also be stored in a portable storage medium such as a disk device connected to a packet transmission device, a flexible disk, a CD-ROM, or the like, and can be installed in a computer at the time of execution.

Fourteenth Embodiment

Next, an embodiment in which a packet requiring high reliability and a packet that does not need high reliability is specified with reference to the packet header information, and a packet is transmitted in multiple paths only for a packet requiring high reliability. The technique of the present embodiment is applicable to a network of users whose demand for reliability is various.

In the present embodiment, in the packet transmission apparatus of the transmission side, by referring to a part of the packet header of the packet, a packet that requires high reliability and a packet that does not need high reliability are distinguished, and a packet requiring high reliability (hereinafter referred to as "highly reliable packet") is duplicated. It transmits on all independent paths and does not duplicate packets that do not need high reliability, and transmits only one of the independent paths. The packet transmission apparatus on the reception side determines whether or not the packet has arrived in an independent path by referring to a part of the packet header at the reception unit to determine whether it is a high reliability packet. With respect to the high reliability packet, further, by determining the identity of the packet data arriving in a plurality of paths, when the same packet arrives in two or more paths, only one is transmitted downstream and the other is discarded. If the same packet arrives on only one path, the packet is sent downstream.

30 shows an example of the configuration of a transmission function unit of a packet transmission device in this embodiment. This configuration is a configuration in which the high reliability judging section 116 is provided between the receiving section 111 and the copy section 112 in the configuration shown in FIG. 8. In the configuration of FIG. 30, a copy and an identifier are given to a packet determined to be a packet requiring high reliability by the high reliability judging unit 116. If the high reliability determination unit 116 determines that the packet does not require high reliability, the packet is transmitted from one of the transmission unit 114 ₁ and the transmission unit 114 ₂ without performing copying or identifier assignment. It is. As shown in Fig. 31, the configuration may be copied after the identifier is assigned to the high reliability packet. After copying, it may be determined whether or not it is a high reliability packet, and if it is not a high reliability packet, only one of the copied packets may be transmitted.

The configuration including the high reliability judging section in the transmission function section and the function of judging whether or not the reception section in the reception section has a high reliability packet is applicable to the packet transmission apparatus of another embodiment.

In the high reliability judging unit 116, a method for distinguishing a packet requiring high reliability from a packet not requiring high reliability includes, for example, the following method.

If the target packet is an Ethernet packet, the arrival port number to the previous step switch of the packet transmitter, the type value of the layer 3 protocol in the packet header, the destination MAC address in the frame header, the source MAC address, and the 802.1Q VLAN. Use one of priority (CoS value) included in tag, VLAN-ID, DiffServ code point value (ToS value) included in IP header, destination port number of UDP or TCP, source port number of UDP or TCP Thus, it can be determined whether or not the packet requires high reliability.

If the target packet is a packet corresponding to the MPLS, it can be determined whether or not the packet needs high reliability by using any one of a destination MAC address, a source MAC address, or a CoS value (Exp value) of a shim header.

Also in the case other than the above, if the target packet includes the IP packet, it can be determined whether or not the packet needs high reliability by using any one of the ToS value, the source IP address, and the destination IP address of the IP packet. .

Then, in the packet transmission apparatus on the transmitting side, when assigning a path identifier, it may include a path identifier reflecting the priority of the user packet. When a plurality of path identifiers are included, one or more of the path identifiers may include path identifiers reflecting the priority of the user packet. Specifically, for example, in the case of using a VLAN tag or shim header for the path identifier, the CoS value of the user-provided VLAN tag or the CoS value (Exp value) of the user-provided shim header or the ToS value of the user IP header Is prioritized, and the value is reflected in the path identifier to insert.

32 shows an example in which the priority of the user packet is reflected in the path identifier to be inserted.

(a) refers to a CoS value of a user VLAN tag (type value 8100) in the packet transmission apparatus on the transmitting side, and reflects the value as a CoS value of a path identifier (VLAN tag of type value 9100) newly assigned. An example is shown. (b) shows an example in which the transmitting side packet transmitter refers to a ToS value in an IP header of a user packet and reflects it as a CoS value of a path identifier (VLAN tag of type value 9100) to which the value is newly given. It is shown. (c) refers to the tag of the inner side (the side near the data) of the two path identifiers which refer to the CoS value of the user VLAN tag (type value 8100) in the packet transmission apparatus of the transmitting side and give the value newly; The example reflected to the CoS value of VLAN tag of type value 8100 is shown. (d) refers to the ToS value in the IP header of the user packet in the packet transmission apparatus on the transmitting side, and among the two newly provided path identifiers, the value is set to the CoS of the inner tag (VLAN tag of the type value 8100). The example reflected as a value is shown.

By reflecting the priority to the path identifier to be inserted, it becomes possible to perform priority control according to the priority in the switch between the packet transmission apparatuses. Further, in the packet transmission apparatus, the ToS control in the switch may be unnecessary by performing the conversion from ToS to CoS.

       In addition, the priority may be reflected in the field of the path identifier and may be reflected in the field of the sequence identifier (counter) and the field of the timestamp to be described later. That is, the format of the newly inserted sequence identifier or time stamp is the same format as the VLAN tag of the 802.1Q specification, and the sequence information or time information is described in the VLAN-ID field of the VLAN tag. The CoS value of the user packet is reflected as the CoS value of the VLAN tag.

An example is shown in FIG. In the example shown in FIG. 33, the form of a newly inserted sequence identifier or timestamp conforms to the VLAN format of 802.1Q, gives 8100 indicating a VLAN tag as a type value, and refers to a user packet as a CoS value. The case where the value obtained by this is given is shown.

In addition, as shown in FIG. 6 and the like, when one or more VLAN tags or shim headers are assigned to a packet, a VLAN tag or shim header provided to the innermost side of the VLAN tag or shim header is closer to the data. The order identifier or a timestamp to be described later may be inserted further inside the. In this way, in the packet transmission apparatus, it is possible to first perform control with the relay switch without copying the sequence identifier CoS value or the like. In other words, the number of tags to be assigned is reduced. In this case, the reception side determines the read position of the sequence identifier or timestamp in accordance with the insertion position. Fig. 34 shows an example of inserting a sequence identifier (counter) inside the user when the user attaches one VLAN tag.

The method of high reliability judgment, the method of reflecting the priority, etc. which were demonstrated in this embodiment are applicable to each other embodiment.

[Fifteenth Embodiment]

35 shows a system configuration according to the present embodiment. As shown in Fig. 35, in the system according to the present embodiment, the packet transmitting apparatus 15 on the transmitting side is provided with a plurality of receiving side packet transmitting apparatuses 16, through the wide area network a21 and the wide area network b22. 17, 18) to form a point-multipoint network.

In this embodiment, when the independent path is two series, the structure of the transmission function part of a transmitter side apparatus can employ | adopt the same structure as the structure shown in FIG. That is, the receiver 111 receives the packet from the user network, duplicates the packet in the copy unit 112, and assigns a path identifier and an order identifier to each of the two packets after the copy, and the two paths on the wide area network side. Send it out. As shown in Fig. 36, the packet may be duplicated and transmitted in two paths after the path identifier and the order identifier are assigned.

37 shows an example of a packet configuration when the target network is Ethernet in the present embodiment. In wide area networks, identifiers are assigned using VLAN technology so that packets can be transmitted over existing Ethernet. In particular, since the path identifier inserted after the MAC address value is referred to during wide area network transmission, it is preferable to assign the path identifier based on the existing VLAN technology (IEEE 802.1Q). Further, the sequence identifier indicating the sequence information may be a tag (4 bytes) conforming to the VLAN description (IEEE 802.1Q), and the byte length is 4 bytes, but does not have to conform to the VLAN description (IEEE802.1Q). You may.

38 is a configuration diagram of the reception function unit 320 of the packet transmission apparatus 300 according to the present embodiment. 38 does not show the transmission function unit. As illustrated in FIG. 38, the reception function unit 320 includes a reception unit 321 ₁ and 321 ₂ for receiving a packet in each path, a memory A 322 _A for temporarily storing a packet, and a memory B 322 _B. ), A control unit 326 for controlling packet transmission and discarding with reference to the order information, and a transmitting unit 325 for transmitting the packet to the user network side. The operation of the reception function unit 320 according to the present embodiment is as follows.

On the wide area network side, the receivers 321 ₁ and 321 ₂ receive packets via two paths, and store the received packets in the memory A 322 _A and the memory B 322 _B, respectively. The controller 326 refers to the oldest packets (the earliest arrived packets) of the memory A 322 _A and the memory B 322 _B , respectively, and transmits the old packets first downstream by comparing the sequence identifiers. do. The control procedure by the control unit 326 will be described with reference to the flowchart in FIG. 39.

The controller 326 refers to the presence or absence of packets in the memory A 322 _A and the memory B 322 _B , and conditionally branches into two systems when there are packets in both memories and when there is a packet in either memory. (Step 601).

(1-1) When there are packets in the two memories, the sequence identifiers CA and CB of the packets of A and B systems are read out (step 602), and the numbers are compared (step 603). Conditional branching is performed for CA <CB, for CA = CB, and for CA> CB.

(1-1-1) In the case of CA <CA, the packet of system A is sent downstream, the packet is removed from the memory A, and the packet of system B is in the standby state, and the packet of the both system is referred to. Return to the state (step 604).

(1-1-2) In case of CA = CB, the A-type (or B-type) packet is transmitted downstream, the transmitted packet is removed from the memory, and the B-type (or A-type) packet is discarded. It removes from memory and returns to the packet reference state of the poultry (step 605).

(1-1-3) In the case of CA> CB, the packet of system B is sent downstream, the packet is removed from the memory, and the packet of system A is returned to the poultry packet reference state while waiting (step A). 606).

That is, a packet with a small number is judged to be a packet to be transmitted next, is sent downstream, and is removed from the memory. The packet with the higher number waits and returns to the packet reference waiting of the chicken.

(1-2) In step 601, when there is a packet in only one of A and B, the process branches to the case where there is a packet only in the A system and the case where there is a packet only in the B system (step 607).

(1-2-1) If there is a packet only in system A, the packet waiting time of system B is checked (step 608).

(1-2-1-1) If the packet waiting time of system B has not expired, the packet reference state of both systems is returned.

(1-2-1-2) When the packet waiting time of system B has expired, the packet of system A is transmitted downstream, removed from memory A, and the packet is returned to reference status (step 609).

(1-2-2) In step 607, if there is a packet only in system B, the packet waiting time of system A is checked (step 610).

(1-2-2-1) If the packet waiting time of system A has not expired, the flow returns to the packet reference state of both systems.

(1-2-2-2) When the packet waiting time of system A has expired, the packet of system B is sent downstream, removed from memory B, and the packet is returned to the reference state of the farm (step 611).

40 is another flowchart according to the present embodiment. FIG. 40 shows a procedure for transmitting a packet once downstream without performing packet waiting of the A system and the B system in the processes subsequent to (1-2) in FIG. 39.

In this embodiment, since the reception function of the packet transmission apparatus does not manage the order of the transmitted packets, when the arrival time difference of the same packet is large in the A system and the B system (more than the packet waiting time), There is a possibility of sending the same packet downstream. Therefore, it is necessary to pay attention to the path delay setting.

[16th Embodiment]

41 shows the configuration of the reception function unit 420 of the packet transmission apparatus 400 of the present embodiment. The transmission function unit is not shown. As shown in FIG. 41, the reception function unit 420 of this packet transmission apparatus is a counter unit 427 which manages the order after completion of the transmission in the configuration of the reception function unit of the packet transmission apparatus according to the fifteenth embodiment. It is equipped with more.

The control unit 426 compares the sequence identifiers of the oldest packets (the earliest arrived packets) of the memories A and B with the counter values of the counter unit 427 indicating the sequence numbers of the already transmitted packets. Determine the unsent packet to send next downstream. The processing procedure by the control unit 426 will be described with reference to the flowchart in FIG. 42.

As shown in FIG. 42, the control unit 426 checks the presence or absence of packets in the memory A 422 _A and the memory B 422 _B , and the packet exists in either of the two memories. If there is, condition branching to two systems (step 701).

(2-1) When there are packets on both sides, the sequence identifiers CA and CB of the packets of A and B systems are read out (step 702), and the magnitudes of the counter values CF are compared with each other. First, regarding A, conditional branching is performed for CF≥CA and for CF <CA (step 703).

(2-1-1) In the case of CF≥CA, the packet of system A is discarded, the packet is removed from the memory A, and the process moves to the comparison of CB and CF (steps 704 and 705).

(2-1-1-1) In the case of CF≥CB, the packet of system B is also discarded, removed from memory B, and returns to the poultry packet reference state (step 706).

(2-1-1-2) In the case of CF <CB, the packet of system B is in a waiting state, and returns to the packet reference state of both systems (step 707).

(2-1-2) In step 703, in the case of CF <CA, the flow moves to the comparison of CB and CF while waiting for the A packet waiting state (step 708).

(2-1-2-1) In case of CF≥CB, the system B packet is discarded, removed from the memory B, and returned to the poultry packet reference state while the system A packet is waiting (step 709).

(2-1-2-2) In step 708, in the case of CF <CB, the A-type and B-type packets are moved to the comparison between CA and CB while waiting for the A-type and B-type packets (step 710).

(2-1-2-2-1) In the case of CA <CB, after transmitting the packet A downstream, removing the packet from memory, and resetting the counter value CF = CA, the packet B is waiting. In return, it returns to the poultry packet reference state (step 711).

(2-1-2-2-2) In case of CA = CB, transmits the A-type (or B-type) packet downstream, removes the transmitted packet from memory, and resets the counter value CF = CA (CB). After that, the B-B packet is discarded. In other words, it is removed from memory (step 712). Then, it returns to the poultry packet reference state.

(2-1-2-2-3) In the case of CA> CB, after sending the B packet downstream, removing the packet from memory, and resetting the counter value CF = CB, the A packet is in a waiting state. It returns to the poultry packet reference state (step 713).

(2-2) In step 701, when there is a packet in only one of A and B, the process branches to the case where there is a packet only in the A system and the case where there is a packet only in the B system (step 714).

(2-2-1) If there is a packet only in system A, the packet waiting time of system B is checked (step 715).

(2-2-1-1) If the packet waiting time of system B has not expired, the packet reference state of both systems is returned.

(2-2-1-2) When the packet waiting time of system B has expired, the sequence identifier CA of the packet A is read, and the process moves to comparison with the CF (steps 716 and 717).

(2-2-1-2-1) In case of CF≥CA, the A system packet is discarded by removing it from the memory A, and the process returns to the poultry packet reference state (step 718).

(2-2-1-2-2) In the case of CF <CA, the A packet is sent downstream, the packet is removed from the memory, the counter value is reset to CF = CA, and the packet is returned to the reference status of the poultry packet. Go (step 719).

(2-2-2) In step 714, if there is a packet only in system B, the packet waiting time of system A is checked (step 720).

(2-2-2-1) If the packet waiting time of the A system has not expired, the flow returns to the poultry packet reference state.

(2-2-2-2) When the packet waiting time of system A has expired, the sequence identifier CB of system B packets is read out and the process moves to comparison with CF (steps 721 and 722).

(2-2-1-2-1) In the case of CF≥CB, the B system packet is discarded by removing it from the memory B, and it returns to the poultry packet reference state (step 723).

(2-2-1-2-2) In case of CF <CB, transfer the system B packet downstream, remove the packet from memory B, reset the counter value CF = CB, and return to the poultry packet reference state. Return (step 724).

In this embodiment, the size of the counter value CF which memorize | stores the order of the packet already transmitted, and the sequence identifier C1-Cn of the packet of the target n system (n = 2 in this embodiment) are compared. . The packet which is larger than CF among C1 to Cn and which has the smallest order among C1 to Cn (when there are a plurality of smallest order packets) is used as the next packet to be transmitted.

In this embodiment, since the order of the transmitted packets is managed by using the counter value CF, even when the arrival time difference of the same packet is large in the A system and the B system (more than the packet waiting time), a plurality of identical packets It is possible to realize non-disruption without transmitting downstream.

As an identifier for identifying the identity between packets, a timestamp may be used in addition to the above sequence identifier. In this case, the packet transmitter on the transmitting side attaches a time stamp to the target packet. In the packet transmission apparatus on the receiving side, time stamps arriving from a plurality of systems are compared and the order is identified.

Also, as in the present embodiment, the identification of the transmitted packet on the receiving side is performed by storing the time stamp preceding m (m is an integer of 1 or more) in the latest packet, and storing the stored time stamp and the next arrived packet. By comparing the timestamps of &lt; RTI ID = 0.0 &gt;, &lt; / RTI &gt;

In this case, the reception function unit of the packet transmission apparatus is provided with a memory for storing the time stamp, as shown in FIG. As this memory, RAM or CAM can be used. In the case of RAM, the data to be compared is compared with each of the data stored in the memory to determine the identity. In the case of CAM, the data to be compared can be collectively compared in a data list in the memory, so that high speed judgment is possible.

In addition to giving the packet an identifier for determining the identity, the receiver may determine the identity between packets based on a value obtained by applying a function that has already been determined for the arrived packet. As a function, for example, it is possible to use a predetermined hash function.

In this case, in performing the identification of the transmitted packet, as shown in FIG. 44, the same memory as in FIG. 43 is maintained. 43, it is possible to use RAM, CAM, and the like as the memory.

On the receiving side, the method described in this embodiment for determining whether or not the arrived packet is a transmitted packet can also be applied to other embodiments.

[17th Embodiment]

Although the structure of the reception function part which concerns on 17th Embodiment is the same as that of the structure of 16th Embodiment shown in FIG. 41, a process operation differs.

In the present embodiment, among the n independent paths (n = 2 in the present embodiment), the path in which the packet arrived soonest after the start of communication is used as the current system, and the order of the transmitted packets is determined. Compare the counter value CF to be stored with the size of the sequence identifier (Ci: 1≤i≤n) of the packets of the series employed as the current system. send. In addition, when the arrival of a packet in the current system is lost, another route in which the packet arrives quickly is adopted as a new current system and the packet is transmitted downstream.

In the seventeenth embodiment, the processing operation will be described using the flowcharts of FIGS. 45 and 46.

In the method shown in Fig. 45, an example in which the packets arrive quickly among the A system and the B system is set to 0 system and transmitted (the other system is 1 system). If the 0 system is revived, the method returns to the 0 system. The flowchart of FIG. 45 is demonstrated below.

First, the process conditionally branches to whether the packet arrives quickly in either the A system or the B system (step 801).

(3-1) When packet A arrives early or when A and B arrive simultaneously, the sequence identifier CA of system A is read (step 802), and large comparison is made with the counter value CF (step 803). ). Conditional branching for CF≥CA and CF <CA.

(3-1-1) In the case of CF <CA, a packet of system A is transmitted, the packet is removed from memory A, CF is set to CF = CA, and if there is a packet of system B, the oldest one is removed ( Step 804), the flow advances to confirming the presence or absence of a packet in the memory A (step 805).

(3-1-1-1) In step 805, when there are packets in the memory A, the packet processing returns to the order reading order.

(3-1-1-2) In step 805, when there is no packet in the memory A, the flow advances to checking the presence or absence of a packet in the memory B (step 806).

(3-1-1-2-1) In step 806, when there is a packet in the memory B, the sequence identifier CB of the B system is read (step 807), and compared with the counter value CF in large ( Step 808). Conditional branching is performed for CF≥CB and CF <CB.

(3-1-1-2-1-1) If CF≥CB, the packet of system B is discarded, and after removing the packet from memory B (step 809), the flow returns to the confirmation of the presence or absence of a packet in memory A.

(3-1-1-2-1-2) If CF <CB, transmit the B-type packet, remove the packet from the memory, set CF to CF = CB (step 810), then check whether the packet exists in memory A. Return to OK.

(3-1-1-2-2) In step 806, when there are no packets in the memory B, the process returns to confirming the early arrival of the A and B systems.

(3-1-2) In step 803, in the case of CF≥CA, the A system packet is discarded, the packet is removed from the memory A (step 811), and the flow returns to the confirmation of the presence or absence of the packet in the memory A. Thereafter, the same processing as in steps 805 to 810 is executed.

(3-2) In step 801, when the packet B arrives quickly, the sequence identifier CB of system B is read (step 812), and compared with the counter value CF (step 813). Conditional branching is performed for CF≥CB and CF <CB.

(3-2-1) In the case of CF <CB, send a packet of B system, remove the packet from memory B, set CF = CB, and remove the oldest one if there is a packet of A system (step 814). In step 815, the flow advances to the confirmation of the presence or absence of the packet in the memory B.

(3-2-1-1) In step 815, when there are packets in the memory B, the packet returns to the order reading order.

(3-2-1-2) In step 815, when there is no packet in memory B, the flow returns to the confirmation of the presence or absence of a packet in memory A (step 816).

(3-2-1-2-1) In step 816, when there is a packet in the memory A, the sequence identifier CA of the A system is read (step 817), and large comparison with the counter value CF ( Step 818). Conditional branching for CF≥CA and CF <CA.

(3-2-1-2-1-1) If CF≥CA, the A system packet is removed from the memory and discarded (step 819), and the flow returns to the confirmation of the presence or absence of the packet in the memory B.

(3-2-1-2-1-2) If CF <CA, transmit the A-type packet, remove the packet from memory A, set CF to CF = CA (step 820), and then store the packet in memory B. Return to the confirmation of presence.

(3-2-1-2-2) In step 816, when there is no packet in the memory A, the process returns to the step of confirming early arrival of the A and B systems.

(3-2-2) In step 813, in the case of CF≥CB, the B system packet is discarded and removed from the memory B (step 821), and the flow returns to the confirmation of the presence or absence of the packet in the memory B. Thereafter, the same processing as in steps 815 to 820 is performed.

In the method shown in Fig. 46, the side in which the packets arrive early among the A and B systems is transmitted as the 0 system (the other side is the 1 system), and when the 0 system is stopped, the packet of the 1 system is transmitted. It is a method of using the first-order packet until the first-system loses communication due to failure or construction, regardless of whether the zero-system is revived or not.

In the flowchart of FIG. 45, when the system A arrives early, the flow chart of FIG. 45 is followed by step 809 or step 810, instead of returning to the confirmation of the presence or absence of the packet of memory A (step 805), the packet of memory B is returned. Confirmation of presence or absence ”. In the case where the system B arrives early, in the flowchart of FIG. 45, after step 820 or step 819, instead of returning to the confirmation of the presence or absence of the packet in the memory B (step 815), the confirmation of the presence or absence of the packet in the memory A is performed. Return to 」.

45, 46, the processing of steps 805, 815, 905, and 915 may be moved to the next step for the first time only after the waiting time timer expires. As a result, the frequency of switching the system can be reduced.

[Eighteenth Embodiment]

47, the reception function part 520 which concerns on 18th Embodiment shows a block diagram. This configuration is the same as that in the fifteenth embodiment shown in FIG. 38, but in the present embodiment, the control unit 526 does not refer to the packet order identifier.

48 is a flowchart of the operation. In the present embodiment, a packet is transmitted as soon as the sequence of arrival of the packet is 0 system and the 0 system is the current system (steps 1002 to 1003 and steps 1007 to 1008). In accordance with the presence or absence of the packet stored in the memory of the reception function unit 520 and the expiration and completion of the packet waiting time, switching determination between 0 and 1 systems is performed (steps 1004 to 1005 and 1009-1010). FIG. 48 shows a case where the current system does not return to the 0 system unless the first system is disconnected after switching the current system from the 0 system to the 1 system. Another example is shown in FIG. The example of FIG. 49 shows a case where the current system is returned to its original state when the 0 system is restored after converting from the current system 0 to the first system.

[19th Embodiment]

Next, a nineteenth embodiment will be described. 50, the network block diagram which concerns on this embodiment is shown. As shown in Fig. 50, this network is a multipoint-multipoint network for transmitting packets from a plurality of transmission sources to a plurality of destinations.

Fig. 51 shows an example packet configuration of Ethernet according to the present embodiment. As shown in Fig. 51A, a VLAN tag (4 bytes) and sequence information (4 bytes) corresponding to a path for transmitting a packet and a sender are added. The configuration shown in Fig. 51B may be used. In the case of Fig. 51B, a 4-byte VLAN tag is added as the route information, and an ID number for distinguishing a transmission source is assigned in the sequence identifier.

52 is a configuration diagram of the reception function unit 620 of the packet transmission device 600 in the present embodiment. The transmission function unit is not shown. In this embodiment, since sequence identifiers of different sequences are provided for each transmission source, each transmission source includes a configuration for realizing the functions shown in the embodiments described above. That is, as shown in Fig. 52, the reception function unit 620 of this embodiment includes a memory A 622 _A , a memory B 622 _B , a control unit 626, and a counter unit 627 for each transmission source. Doing. In addition, a control unit 628 is provided, and the control unit 628 confirms transmission source information, and distinguishes a storage destination for the memory for each transmission source.

[20th Embodiment]

53 shows another example of the reception function unit 720 of the packet transmission apparatus 700 applicable to the network shown in FIG.

In the configuration shown in Fig. 53, the memories A and B storing the reception packets are not distinguished for each transmission source. That is, one memory A 722 _A and one memory B 722 _B are included. In the present embodiment, the control unit 728 obtains CA and CB values and source information from the packets of the memory A 722 _A and the memory B 722 _B. Based on the source information, the controller 728 obtains CA and CB values. It transmits to the control part 726 of the said transmission source. The control unit 726 of the sender performs order comparison of the packets, and the result of the order comparison is sent to the control unit 728. The control unit 728 discards, transmits, and queues the packets to the memory A 722 _A and the memory B 722 _B based on the order comparison result. The content of processing for one transmission source is the same as the embodiment described so far.

[21st Embodiment]

In the above embodiment, when the counter value is used as the sequence identifier for determining the identity of the packet, since the counter value is finite (returns to zero after reaching the maximum value), the counter value after once returning to zero, It is not possible to identify the counter value before returning to zero. In this embodiment, it is managed whether or not it has returned to zero, and the counter value is compared in consideration of that.

54 shows the configuration of the reception function unit 820 of the packet transmission apparatus 800 according to the present embodiment. As shown in FIG. 54, the reception function unit 820 includes, in addition to the configuration shown in FIG. 41, an arrival counter / circulation counter 829 in the A and B systems, respectively.

Referring to Fig. 55, the operation of the apparatus will be described.

In A system and B system, weekly meetings are managed independently. The following describes the circumferential management of the A system, but the same applies to the B system.

The A-system, if a packet of the sequence n ₁ arrives, the destination counter value of the A-system to n _1, and resetting the state-time timer of the A-system of the controller 826 (of FIG. 55 (1), and then packets of the Until this arrival, when the time timer counts the time corresponding to n ₁ to the maximum value of the sequence number, 1 is added to the time counter of the A system (Fig. 55 (2)). If the maximum value of the sequence number is N _max , it is the time corresponding to ((minimum frame length + inter-frame interval)) x (N _max -n ₁ ). When this arrives, the sequence number of the packet is used as it is for sequence comparison.

When the next packet (order n ₂ ) of system A arrives within the delay time difference between system A and system B (FIG. 55 (3)), the turn counter becomes +1, and if n ₂ <n ₁ , The order of the packets of the A system is set to n ₂ + N _max and used for order comparison. Even if the winding counter is in a state of +1, if n ₂ ≥ n ₁ , order comparison is performed using n ₂ . After (2), if the next packet does not arrive within the delay difference time between the A and B systems, the completed counter value returns to zero.

In the above process, when the circumferential counters in the A and B systems coincide, both circumferential counters are turned to zero. Further, the maximum value N _max in the above-described order is set such that delay difference ≤ ((time corresponding to minimum frame length (64 bytes) + inter-frame interval)) x N _max between A and B.

In addition, by recording the number of times the counter value returned to 0 at the transmitting side, the number of times the counter value returned to 0 at the receiving side may be used for comparison. Fig. 56 shows an example of the packet configuration in the case of Ethernet when this method is adopted. As shown in Fig. 56, in addition to the counter value, the number of times the counter has returned to zero is recorded. On the receiving side, the number of times and the counter value are compared between packets. In other words, when the number of times the counter value returns to zero is the same, the counter values are compared.

The present invention is not limited to the above-described embodiments, and various modifications and applications are possible within the scope of the claims.

Claims (74)

In a network apparatus for transmitting a packet, as a packet transmission method, The apparatus on the transmitting side creates two copies of the transmission packet, assigns each of the copied packets a sequence number for identifying the same transmission order, and further, an identifier according to the transmission / reception pair to each of the copied packets. To send, The receiving device receives the packet at each of two receiving means, Confirm an identifier according to the transmission / reception pair, If the identifiers match, identify the packet and the sequence including the same information based on the sequence number, A packet transmission method, characterized in that one of the packets of the same order is selected and sent out downstream, and the other packets are discarded, and when only one side arrives, the arrived packet is transmitted downstream. In a network apparatus for transmitting a packet, as a packet transmission method, The apparatus on the transmitting side creates two copies of the transmission packet, assigns each of the copied packets a sequence number for identifying the same transmission order, and further, an identifier according to the transmission / reception pair to each of the copied packets. And assigns and transmits an identifier according to a path, The apparatus at the transmitting side receives the packets at two receiving means, respectively, Identifying an identifier according to the transmission / reception pair and an identifier according to the path, If the identifiers match, the packets including the same information and their order are identified based on the sequence numbers. A packet transmission method, characterized in that one of the packets of the same order is selected and sent out downstream, and the other packets are discarded. If only one side arrives, the arrived packet is transmitted downstream. The packet transmission according to claim 1 or 2, wherein the receiving side apparatus temporarily stores packets transmitted in two paths in two FIFO memories, and selects packets transmitted normally and transmits them downstream. Way. The packet transmission according to claim 1 or 2, wherein the receiving side apparatus temporarily stores packets transmitted in two paths in two circular hash memories, and selects packets transmitted normally and transmits them downstream. Way. 3. The receiving apparatus according to claim 1 or 2, wherein the receiving side apparatus temporarily stores packets transmitted in two paths in two FIFO memories, and uses a third memory shared by two receiving means as a cyclic hash. Selecting a packet transmitted normally and transmitting the packet downstream. The packet transmission method according to claim 1 or 2, wherein Ethernet is used as a packet transmission technique. 7. The packet transmission method according to claim 6, wherein a tag field and a counter field are inserted after a source MAC address of an Ethernet packet, and a VLAN tag and a sequence number along a path are described. The packet transmission method according to claim 6, wherein a tag field and a counter field are inserted after the source MAC address of the Ethernet packet to describe a VLAN tag and a sequence number according to a transmission / reception pair. The packet transmission method according to claim 6, wherein a tag field and a counter field are inserted after the source MAC address of the Ethernet packet, and a VLAN tag corresponding to the transmission / reception pair and the transmission path and a sequence number are described. The packet transmission method according to claim 6, wherein a tag field and a counter field are inserted after the source MAC address of the Ethernet packet, and a VLAN tag according to a transmission path, an identification ID and a sequence number according to a transmission / reception pair are described. The packet transmission method according to claim 1 or 2, wherein MPLS is used as a packet transmission technique. The packet transmission method according to claim 11, wherein a tag field and a counter field are inserted before a shim header of the MPLS, and a shim header and a sequence number according to a transmission path are described. 12. The packet transmission method according to claim 11, wherein a tag field and a counter field are inserted before the MPLS shim header, and a shim header and a sequence number according to a transmission / reception pair are described. The packet transmission method according to claim 11, wherein a tag field and a counter field are inserted before a shim header of the MPLS, and a shim header according to a transmission path, an identification ID and a sequence number according to a transmission / reception pair are described. The packet transmission method according to claim 11, wherein a tag field and a counter field are inserted before a shim header of the MPLS, and a shim header and a sequence number corresponding to a transmission path and a transmission / reception pair are described. The memory region using a cyclic hash is divided into n (n is an integer of 2 or more), and addresses 1 to n are given. When the receiving device receives a packet, the packet is stored in a memory area whose address is the remaining value when the counter value is divided by n, and the packet having the counter value N is preceded by the packet having the counter value Nn. Even when it arrives, once the packet is stored in a memory area whose address is the remaining value obtained by dividing the counter value N by n, and when the packet is read, the packet is read in the order of the counter value of the packet. The packet transmission method of correcting the reverse order of arrival order within the first order in the correct order at the time of reading. The packet transmission method according to claim 1 or 2, wherein the packet transmission technique uses a packet transmission technique using encapsulation of a variable length packet. 18. The packet transmission method according to claim 17, wherein when a header for encapsulation is given to the variable length packet, a counter field is inserted after the header for encapsulation, and a sequence number is described. The packet transmission method according to claim 18, wherein, in the receiving device, an identifier according to a transmission / reception pair or an identifier along a path is extracted from a header for encapsulation. Packet transmission device for transmitting a packet, A transmitting function means and a receiving function means, The transmitting function means is: Copying means for making two copies of a transmission packet; Number / identifier assigning means for assigning a sequence number for identifying the same transmission order to each of the packets copied by the copying means, and further assigning an identifier according to the transmission / reception pair to each of the copied packets; And Packet transmission means for transmitting two packets to which the sequence number and the identifier are assigned; The receiving function means is: Packet receiving means for receiving each of the two packets transmitted from the transmitting function means; A memory for storing each of the two received packets; Read the two packets stored in the memory, recognize the identifier according to the transmission / reception pair, and if the identifiers match, identify the packet including the same information and the sequence based on the sequence number, Selection means for selecting one of; And And transmitting means for transmitting the packet selected by the selecting means downstream, and discarding the other packet, wherein, if only one piece arrives, transmitting means for transmitting the arrived packet downstream. Packet transmission device for transmitting a packet, A transmitting function means and a receiving function means, The transmitting function means is: Copying means for making two copies of a transmission packet; A sequence number for identifying the same transmission order is assigned to each of the packets copied by the copying means, and further, a number and identifier for assigning an identifier according to a transmission / reception pair and an identifier along a path to each of the copied packets. Grant means; Packet transmitting means for transmitting two packets to which the sequence number and the identifier are assigned; The receiving function means is: Packet receiving means for receiving each of the two packets transmitted from the transmitting function means; A memory for storing each of the two received packets; Read the two packets stored in the memory, recognize the identifier according to the transmission / reception pair and the identifier according to the path, and if the identifiers match, the packet having the same information based on the sequence number Selecting means for identifying a sequence and a sequence thereof, and selecting one of the packets of the same sequence; And And transmitting means for transmitting the packet selected by the selecting means downstream, discarding another packet, and transmitting the arrived packet downstream when only one piece arrives. 22. The apparatus of claim 20 or 21, wherein the memory is a FIFO memory, and wherein the selecting means of the receiving function means includes means for selecting a packet normally transmitted from a packet temporarily stored in two FIFO memories. Packet transmission device. 22. The apparatus of claim 20 or 21, wherein the memory is a cyclic hash memory, and wherein the selecting means of the receiving function means includes means for selecting a normally transmitted packet from packets temporarily stored in two cyclic hash memories. , Packet transmission device. 22. The apparatus according to claim 20 or 21, wherein the memory is a FIFO memory, and the receiving function means further comprises a third memory shared by the two packet receiving means, and uses the third memory as a cyclic hash. Means for selecting a packet transmitted normally. The packet transmission apparatus according to claim 20 or 21, wherein said transmission function means and said reception function means use Ethernet as a packet transmission technique. The packet transmission apparatus according to claim 25, wherein a tag field and a counter field are inserted after a source MAC address of an Ethernet packet, and a VLAN tag and a sequence number along a path are described. The packet transmission apparatus according to claim 25, wherein a tag field and a counter field are inserted in a source MAC address of an Ethernet packet, and a VLAN tag and a sequence number corresponding to a transmission / reception pair are described. The packet transmission apparatus according to claim 25, wherein a tag field and a counter field are inserted in a source MAC address of an Ethernet packet, and a VLAN tag corresponding to a source pair and a transmission path and a sequence number are described. The packet transmission apparatus according to claim 25, wherein a tag field and a counter field are inserted in a transmission source MAC address of an Ethernet packet, and a VALN tag according to a transmission path, an identification ID and a sequence number according to a transmission / reception pair are described. The packet transmission apparatus according to claim 20 or 21, wherein MPLS is used as a packet transmission technique. 31. The packet transmission apparatus according to claim 30, wherein a tag field and a counter field are inserted before a shim header of the MPLS, and a shim header and a sequence number according to a transmission path are described. 31. The packet transmission apparatus according to claim 30, wherein a tag field and a counter field are inserted before a shim header of the MPLS, and a shim header corresponding to a transmission / reception pair and a sequence number are described. 31. The packet transmission apparatus according to claim 30, wherein a tag field and a counter field are inserted before a shim header of the MPLS, and a shim header according to a transmission path, an identification ID and a sequence number according to a transmission / reception pair are described. 31. The packet transmission apparatus according to claim 30, wherein a tag field and a counter field are inserted before a shim header of the MPLS, and a shim header and a sequence number corresponding to a transmission path and a transmission / reception pair are described. The memory area according to claim 25, wherein the memory area using the cyclic hash is divided into n (n is an integer of 2 or more), and addresses 1 to n are given. When a packet is received, the packet is stored in a memory area addressed to the remaining value when the counter value is divided by n, and even when a packet having a counter value of N arrives before a packet having a counter value of Nn. If the counter value N is divided by n, the remaining value is stored in the address area, and when the packet is read, the packet is read in the order of the counter value of the packet. Correcting the packet in the correct order at the time of reading. The packet transmission apparatus according to claim 20 or 21, wherein, as a packet transmission technique, a packet transmission technique using encapsulation of a variable length packet is used. The packet transmission apparatus according to claim 36, wherein when a header for encapsulation is attached to the packet of variable length, a counter field is inserted after the header for encapsulation, and a sequence number is described. The apparatus according to claim 37, wherein an identifier according to a transmission / reception pair or an identifier along a path is extracted from a header for encapsulation. A communication network for transmitting a packet by determining a transmission destination by referring to destination information of a packet, comprising: a packet transmission method executed by a packet transmission apparatus provided on a packet transmission side and a reception side, Two or more independent paths are established between the packet transmitter on the sending side and the packet transmitter on the receiving side. The packet transmission apparatus on the transmitting side inserts information for distinguishing the order of packets in a portion of the packet which is not referred to in determining the destination of the packet, and copies the packet to generate two or more packets, Send each packet to the independent path, The packet transmission apparatus on the receiving side receives the packets via the independent paths respectively, and refers to the information for distinguishing the order of each packet, thereby identifying the packet and the order containing the same information, and identifying the same information. A packet transmission method, characterized in that one of the packets to be sent downstream in order in the order of the packet. The packet transmission apparatus according to claim 39, wherein the packet transmission apparatus on the transmitting side inserts, in the packet, path identification information identifying the independent path as part of destination information for determining a transmission destination in the communication network, A packet transmission method characterized by sending a packet along a path identified by the packet. 41. The transmission apparatus according to claim 39 or 40, wherein the packet transmission apparatus on the transmission side inserts at least transmission source identification information for identifying the packet transmission apparatus on the transmission side into the packet, The packet transmission apparatus of the receiving side identifies a packet including the same information and its order for each transmission source by referring to the transmission source identification information. 40. The method of claim 39, wherein the two or more independent paths are two or more paths that are physically independent or two or more dedicated lines. A communication network for transmitting a packet by determining a transmission destination by referring to destination information of a packet, the packet transmission system including a plurality of packet transmission apparatuses provided through at least two independent paths on a packet transmission side and a reception side. The packet transmission apparatus used as The packet transmission apparatus includes a transmission function means and a reception function means, The transmission function means, Inserting means for inserting information for distinguishing the order of packets into a portion of the packet which is not referred to in determining a transmission destination of the packet; Copying means for copying the packet to generate two or more packets; And Sending means for sending out the respective packets to the independent paths; Equipped with The receiving function means, Receiving means for receiving packets each on the independent path; Identification means for identifying a packet containing the same information and its order by referring to the information for distinguishing the order of each packet; Selecting means for transmitting one of the packets including the same information identified by the identification means downstream in order according to the order of the packets. 44. The transmission means according to claim 43, wherein the insertion means in the transmission function means inserts, in the packet, path identification information identifying the independent path as part of destination information for determining a transmission destination in the communication network. Is a packet transmitting apparatus, characterized in that for transmitting a packet in the path identified by the corresponding path identifier. 45. The apparatus according to claim 43 or 44, wherein the inserting means in the transmitting function means inserts at least source identification information for identifying the packet transmitting apparatus on the transmitting side into the packet, In the receiving function means, the identifying means identifies a packet including the same information for each transmission source and its order by referring to the transmission source identification information. 44. The apparatus of claim 43, wherein the two or more independent paths are two or more paths that are physically independent or two or more dedicated lines. A communication network for transmitting a packet by determining a destination by referring to destination information of a packet, wherein the packet transmission method is executed by a packet transmission apparatus provided at a packet transmission side and a reception side, Two or more independent paths are established between the packet transmitter on the sending side and the packet transmitter on the receiving side. By referring to a part of the packet header of the packet, the packet transmission apparatus on the transmitting side distinguishes between a high-reliability packet and a packet that is not, transmits the high-reliability packet by copying it to all of the independent paths, The packet transmission apparatus on the receiving side determines whether or not the packet has arrived in the independent path by referring to a part of the packet header, and determines whether the packet has arrived in a plurality of paths. If the same packet arrives in two or more paths, only one is sent downstream and the rest is discarded. If the same packet arrives in only one path, the packet is sent downstream. A packet transmission method. 48. The method according to claim 47, wherein when the packet to be transmitted is an Ethernet packet, a part of the packet header includes a port number arriving at a switch in front of the packet transmitter, a type value of a layer 3 protocol in a frame header, and a frame. Destination MAC address in header, Source MAC address, Priority (CoS value) included in 802.1Q VLAN tag, VLAN-ID, DiffServ code point value (ToS value) included in IP header, Destination port number of UDP, Source of transmission Any one of a port number, a destination port number of TCP, and a source port number. When the packet to be transmitted is an MPLS-compatible packet, part of the packet header is any one of a destination MAC address, a source MAC address, and a CoS value (Exp value) of a shim header. When the packet to be transmitted includes an IP packet, part of the packet header is any one of a ToS value, a source IP address, and a destination IP address of the IP packet. 48. The packet transmission apparatus of claim 47, wherein the packet transmission apparatus on the receiving side determines the identity of the packets arriving on the plurality of paths based on a value obtained by operating a predetermined function with respect to the packets arriving on the plurality of paths. A packet transmission method. 48. The apparatus of claim 47, wherein the packet transmission apparatus on the transmitting side inserts a sequence identifier or timestamp into a packet to be transmitted, and the packet transmission apparatus on the receiving side references the sequence identifier or timestamp inserted on the transmitting side. And determining the identity of the packets arriving in the plurality of paths. 51. The packet transmitting apparatus of claim 50, wherein, if at least one VLAN tag or shim header is assigned to the packet, the packet transmission apparatus on the transmitting side further adds the VLAN tag or shim header that is provided at the innermost side of the VLAN tag or shim header. And a sequence transmitter or timestamp inserted therein, and the packet transmission apparatus on the receiving side determines the read position of the sequence identifier or time stamp inserted into the packet according to the insertion position. 51. The transmission apparatus according to claim 50, wherein the format of the sequence identifier or time stamp inserted into the packet by the packet transmitting apparatus on the transmitting side is the same format as the VLAN tag of the 802.1Q specification, and the packet transmitting apparatus on the transmitting side is the VLAN of the VLAN tag. A packet transmission method, characterized by describing sequence information or time information in an ID field. 51. The packet transmission method according to claim 50, wherein a field for describing sequence information or time information in a sequence identifier or a timestamp inserted into a packet by a packet communication apparatus on the transmitting side is of an arbitrary length. 48. The apparatus of claim 47, wherein the packet transmission apparatus on the transmitting side assigns one or more path identifiers to the packet to be transmitted, and reflects the priority given to the packet to the user network among at least one of the path identifiers. Packet transmission method. 55. The apparatus of claim 54, wherein the packet transmission apparatus at the transmitting side, The path identifier is assigned to the packet using a VLAN tag or a shim header. By determining a priority by referring to a part of the packet header, the priority is reflected in the path identifier, If the packet to be transmitted is an Ethernet packet, a part of the packet header may include a port number arriving at a switch in front of the packet transmission device, a type value of a layer 3 protocol in the frame header, a destination MAC address in the frame header, Source MAC address, priority (CoS value) included in the 802.1Q VLAN tag, VLAN-ID, DiffServ code point value (ToS value) included in the IP header, destination port number of UDP, source port number, TCP destination Is either a port number or a source port number. When the packet to be transmitted is an MPLS-compatible packet, part of the packet header is any one of a destination MAC address, a source MAC address, and a CoS value (Exp value) of a shim header. When the packet to be transmitted includes an IP packet, a part of the packet header is any one of a ToS value of the IP packet, a source IP address, and a destination IP address. In a communication network for transmitting a packet by determining a transmission destination with reference to packet destination information, a packet transmission method executed by a packet transmission apparatus provided on a packet transmission side and a reception side, Two or more independent paths are established between the sending packet transmitter and the receiving packet transmitter. The packet transmission apparatus on the transmission side duplicates the packet and sends it out to all of the independent paths, The packet transmission apparatus on the receiving side receives the packets via the independent paths respectively, and identifies the packets containing the same information by referring to the identification identification information of each packet, and any of the packets containing the same information. In the packet transmission method for transmitting an unsent packet downstream, The packet transmission apparatus on the receiving side stores the identity identification information of the packet which has already been transmitted downstream for the packet preceding m pieces (m is an integer greater than or equal to 1) in the latest packet, and stores the identification identification information stored next. And comparing the identity identification information of the arrived packets to determine whether the arrived packets have finished or not transmitted. The packet transmission method according to claim 56, wherein the identity identification information is a value obtained by acting a sequence identifier or time stamp inserted into a packet or a predetermined function on the packet. The packet transmission method according to claim 56, wherein the receiving side packet transmission apparatus uses a content addressable memory (CAM) as a memory for storing the identity identification information. In a communication network for transmitting a packet by determining a transmission destination with reference to packet destination information, a packet transmission method executed by a packet transmission apparatus provided on a packet transmission side and a reception side, Two or more independent paths are established between the packet transmitter on the sending side and the packet transmitter on the receiving side. The packet transmission apparatus on the transmitting side inserts an order identifier for identifying the order of packets in the packet to be transmitted, duplicates the packet, and sends it out to all of the independent paths, The packet transmission apparatus on the receiving side compares the value CF of the sequence identifier of the transmitted packet with the values of the sequence identifiers of the plurality of packets received on the independent paths, and receives the plurality of packets received on the independent paths. At least one packet of the sequence identifier of the packet that is greater than the value CF of the sequence identifier of the already transmitted packet and has the smallest value among the sequence identifiers of the plurality of packets received on the independent path. The packet transmission method, characterized in that for transmitting one of the packets downstream. In a communication network for transmitting a packet by determining a transmission destination by referring to packet destination information, a packet transmission method executed by a packet transmission apparatus provided on a packet transmission side and a reception side, A plurality of independent paths are set between the sending packet transmitter and the receiving packet transmitter; The packet transmitter on the transmitting side inserts a sequence identifier for identifying the order of packets in the packet to be transmitted, duplicates the packet, and sends it out to all of the independent paths, The packet transmission apparatus on the receiving side receives the packets on the independent paths respectively and references the sequence identifiers of the respective packets to identify packets containing the same information and the sequence thereof, and the packets containing the same information. As a packet transmission method for transmitting one of the downstream in the order of the packet, In the packet transmission apparatus on the receiving side, among the independent paths, the path in which the packet arrived first after the start of communication is the current system, and the value (CF) of the sequence identifier of the transmitted packet is received in the current system. Compares the value of the sequence identifier of the packet to be transmitted, and sets a packet having the sequence identifier larger than the value (CF) of the transmitted packet as the next packet to be transmitted; The packet transmission method characterized by employing another system as a new current system when the arrival of the packet of the current system is interrupted for a predetermined time, and transmitting the packet received in the system downstream. A communication network for transmitting a packet by determining a destination by referring to destination information of a packet, comprising: a packet transmission system including a plurality of packet transmission apparatuses provided through at least two independent paths on a packet transmitting side and a receiving side; As the packet transmission apparatus used, The packet transmission apparatus includes a transmission function means and a reception function means, The transmission function means, By referring to a portion of the packet header of the packet, the means for distinguishing between a highly-reliable packet and a non-trusted packet, duplicating the high-reliable packet, and sending it to all of the independent paths, The receiving function means, With respect to the packet arriving on the independent path, it is determined whether or not it is a high reliability packet by referring to a part of the packet header, and for the high reliability packet, the sameness of the packet data arriving on a plurality of paths is determined, and the same packet is Means for transmitting only one downstream and discarding the other when arriving on two or more paths, and transmitting the packet downstream when the same packet arrives on only one path. Transmission device. 62. The method according to claim 61, wherein when the packet to be transmitted is an Ethernet packet, a part of the packet header includes an arrival port number to a switch at the front end of the packet transmitter, a type value of a layer 3 protocol in a frame header, Destination MAC address in the frame header, source MAC address, priority (CoS value) included in the 802.1Q VLAN tag, VLAN-ID, DiffServ code point value (ToS value) included in the IP header, destination port number of UDP, Is one of a source port number, a destination port number of TCP, and a source port number, When the packet to be transmitted is an MPLS-compatible packet, part of the packet header is any one of a destination MAC address, a source MAC address, and a CoS value (Exp value) of a shim header. When the packet to be transmitted includes an IP packet, part of the packet header is any one of a ToS value, a source IP address, and a destination IP address of the IP packet. The packet receiving apparatus according to claim 61, wherein said receiving function means judges the identity of the packet arriving in the plurality of paths based on a value obtained by operating a predetermined function with respect to the packet arriving in the plurality of paths. , Packet transmission device. The packet transmission apparatus according to claim 61, wherein said reception function means judges the identity of packets arriving in a plurality of paths by referring to a sequence identifier or a timestamp inserted at the transmission side. The transmission function means according to claim 64, wherein, if at least one VLAN tag or shim header is given to the packet, the transmitting function means is located further inside of the VLAN tag or shim header provided in the innermost of the VLAN tag or shim header. And inserting a sequence identifier or a timestamp, wherein said receiving function means determines the reading position of the sequence identifier or timestamp inserted into the packet according to the insertion position. The format of the sequence identifier or timestamp in which the transmission function means is inserted in the packet is the same format as the VLAN tag of the 802.1Q specification, and the transmission function means is stored in the VLAN-ID field of the VLAN tag. A packet transmission apparatus characterized by describing sequence identification or time information. The packet transmission apparatus according to claim 64, wherein a field for describing sequence information or time information in a sequence identifier or a timestamp inserted into the packet by the transmitting function means has an arbitrary length. 63. The method according to claim 61, wherein the transmitting function means assigns one or more path identifiers to the packet to be transmitted, and reflects the priority given to the packet to the user network in at least one of them. Packet transmission device. 69. The apparatus according to claim 68, wherein said transmitting function means assigns said path identifier to said packet using a VLAN tag or a shim header, The priority is determined by referring to a part of the packet header, and the priority is reflected in the path identifier. If the packet to be transmitted is an Ethernet packet, a part of the packet header may include a port number arriving at a switch in front of the packet transmitter, a type value of a layer 3 protocol in the frame header, a destination MAC address in the frame header, Source MAC address, priority (CoS value) included in the 802.1Q VLAN tag, VLAN-ID, DiffServ code point value (ToS value) included in the IP header, destination port number of UDP, source port number, TCP destination Is either a port number or a source port number. When the packet to be transmitted is an MPLS-compatible packet, part of the packet header is any one of a destination MAC address, a source MAC address, and a CoS value (Exp value) of a shim header. When the packet to be transmitted includes an IP packet, part of the packet header is any one of a ToS value, a source IP address, and a destination IP address of the IP packet. A communication network for transmitting a packet by determining a transmission destination by referring to destination information of a packet, comprising: a packet transmission including a plurality of packet transmission devices provided through at least two independent paths on a packet transmitting side and a receiving side The packet transmitter used in the system, The packet transmission apparatus includes a transmission function means and a reception function means, The transmission function means includes means for replicating a packet and sending it out in all of the independent paths, The receiving function means, Means for receiving packets each in the independent path; Means for identifying a packet containing the same information by referring to the identity identification information of each packet; The identity identification information of the packet already transmitted downstream is stored for m packets (m is an integer of 1 or more) in the latest packet, and the stored identity identification information is compared with the identity identification information of the next arrival packet. Means for determining whether the corresponding arrival packet is complete or untransmitted; One of the packets containing the same information, characterized in that it comprises means for transmitting an untransmitted packet downstream. The packet transmission apparatus according to claim 70, wherein the identity identification information is a sequence identifier or time stamp inserted into a packet, or a value obtained by operating a predetermined function on the packet. The packet transmission apparatus according to claim 70, wherein said reception function means includes a content addressable memory (CAM) as a memory for storing said identity identification information.  A communication network for transmitting a packet by determining a destination by referring to destination information of a packet, the packet transmission system including a plurality of packet transmission apparatuses provided through at least two independent paths on a packet transmitting side and a receiving side. The packet transmitter used by The packet transmission device includes a transmission function means and a reception function means, The transmitting function means includes means for inserting a sequence identifier for identifying a sequence of packets in a packet to be transmitted, duplicating the packet, and sending it out in all of the independent paths, The reception function means compares a value CF of a sequence identifier of a transmitted packet with values of sequence identifiers of a plurality of packets received on the independent path, and compares the values of the plurality of packets received on the independent path. Of the at least one packet having a value greater than the value of the sequence identifier of the already transmitted packet (CF) among the value of the sequence identifier and having the smallest value among the sequence identifiers of the plurality of packets received on the independent path. And means for transmitting one packet downstream. A communication network for transmitting a packet by determining a destination by referring to destination information of a packet, the packet transmission system including a plurality of packet transmission apparatuses provided through at least two independent paths on a packet transmitting side and a receiving side. The packet transmission apparatus used in the The packet transmission apparatus includes a transmission function means and a reception function means, The transmitting function means includes means for inserting a sequence identifier for identifying the order of packets in a packet to be transmitted, replicating the packet and sending it out through all of the independent paths, The receiving function means receives the packets in the independent paths respectively, and refers to the sequence identifiers of the respective packets, thereby identifying a packet containing the same information and the sequence thereof, and selecting one of the packets containing the same information. Means for sending downstream in order of the packet, The reception function means uses the current path as the path in which the packet arrives first after starting communication among the independent paths, the value CF of the sequence identifier of the transmitted packet, and the packet received in the current system. Compares the value of the sequence identifier of, and sets the packet having the sequence identifier larger than the sequence identifier value CF of the transmitted packet as the next packet to be transmitted. A packet transmission apparatus characterized by employing another system as a new current system when a packet arrives at the current system at a predetermined time and transmitting the received packet downstream.

Images